Gender and Politeness

Gender and Politeness Gender and Politeness challenges the notion that women are necessarily always more polite than men as much of the language and g...
Author: Hilary Baker
3 downloads 2 Views 1MB Size
Gender and Politeness Gender and Politeness challenges the notion that women are necessarily always more polite than men as much of the language and gender literature claims. Sara Mills discusses the complex relations between gender and politeness and argues that although there are circumstances when women speakers, drawing on stereotypes of femininity to guide their behaviour, will appear to be acting in a more polite way than men, there are many circumstances where women will act just as impolitely as men. The book aims to show that politeness and impoliteness are in essence judgements about another’s interventions in an interaction and about that person as whole, and are not simple classifications of particular types of speech. Drawing on the notion of community of practice, Mills examines the way that speakers negotiate with what they perceive to be gendered stereotypes circulating within their particular group. Sara Mills is Research Professor in the School of Cultural Studies, Sheffield Hallam University. She has published in the areas of feminist post-colonial theory and feminist linguistics, and her books include Feminist Stylistics (1995), Discourse (1997), Michel Foucault (2003), and Language and Gender: Interdisciplinary Perspectives, 1994.

Studies in Interactional Sociolinguistics editors Paul Drew, Marjorie Harness Goodwin, John J. Gumperz, Deborah Schiffrin Discourse strategies John J. Gumperz Language and social identity edited by John J. Gumperz The social construction of literacy edited by Jenny Cook-Gumperz Politeness: some universals in language usage Penelope Brown and Stephen C. Levinson Discourse markers Deborah Schiffrin Talking voices Deborah Tannen Conducting interaction Adam Kendon Talk at work edited by Paul Drew and John Heritage Grammar in interaction Cecilia E. Ford Crosstalk and culture in Sino-American communication Linda W. L. Young Aids counselling: institutional interaction and clinical practice ¨ a¨ Anssi Perakyl Prosody in conversation edited by Elizabeth Couper-Kuhlen and Margaret Selting Interaction and grammar edited by Elinor Ochs, Emanuel A. Schegloff and Sandra M. Thompson Credibility in court Marco Jacquemet Interaction and the development of mind A. J. Wootton The news interview Steven Clayman and John Heritage Gender and politeness Sara Mills

Gender and Politeness SARA MILLS Cambridge University Press

CAMBRIDGE UNIVERSITY PRESS Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo Cambridge University Press The Edinburgh Building, Cambridge CB2 8RU, UK Published in the United States of America by Cambridge University Press, New York www.cambridge.org Information on this title: www.cambridge.org/9780521810845 © Sara Mills 2003 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published 2003 A catalogue record for this publication is available from the British Library ISBN 978-0-521-81084-5 hardback ISBN 978-0-521-00919-5 paperback Transferred to digital printing 2007

To Tony, Gabriel, and Francis

Contents

Acknowledgements Introduction

page viii 1

1 Rethinking linguistic interpretation

16

2 Theorising politeness

57

3 Politeness and impoliteness

121

4 Theorising gender

164

5 Gender and politeness

202

Conclusions

238

Bibliography Index

247 266

vii

Acknowledgements

I would like to thank the Arts and Humanities Research Board for enabling me to complete this book by providing me with a research grant in 2000–2001. I am also very grateful to all of the many people who have helped me with this research, either in conversations about it, through providing anecdotal data, by reading drafts of the chapters, or by agreeing to have their conversations recorded and discuss those conversations with me. Tony Brown, Chris Christie, and Clare Walsh have been particularly generous with their time in this respect. I would like to thank the members of the CrossCultural Linguistic Politeness Research Group: both the publishing group Francesca Bargiela, Yi Li, Sandra Harris, Chris Christie, Bethan Davies, Andrew Merrison, Louise Mullany, Corinne Boz, Abdurrahman Hamza, Manana Tevzadze, and Nino Rukhadze, and also the wider group. I am also grateful to postgraduates and staff at Sheffield Hallam University, most notably, Kathryn Longden, Janine Liladhar, Jo Gavins, Jill LeBihan, Keith Green, Dave Hurry, Mary Peace, and also Jane Mardell, Jess Mardell, and Antony Meehan. Jane Sunderland and Lia Litoselliti also gave me very useful feedback on earlier versions of some of the sections of this book. I should like to thank members of the Templeton Reading group: Jan, Elizabeth, Rosemary, Ginette, Brenda, and the members of the Exploring Faith group, principally, Martin, Gaidagh, Judith, Paula, and Jan. I am grateful to those who completed questionnaires, participated in interviews, and allowed me to tape-record their conversations. Thanks are also due to Alexandra Berry, Nicholas Sarra, and Nick Carson. I am very grateful to my editor Andrew Winnard and to the anonymous reviewer of my first draft who made such helpful comments. viii

Introduction

My main aim in this book is to develop a more community-based, discourse-level model of both gender and linguistic politeness and the relation between them. This is in marked contrast to an almost exclusive focus on the individual in most analyses of politeness. At the same time as describing what gender and linguistic politeness are and how they function, my aim is also to question the stability and solidity of these entities. Instead, I see them as processes or acts of evaluation which people perform in conversation. This process model does not mean that I see gender and politeness as ephemeral or without material effects, but rather indicates that I want to move research away from the notion that politeness or gender consist of a range of stable, predictable attributes. At the same time as radically questioning the nature of gender and politeness, I shall also examine the role of stereotype in the process of assessing people’s linguistic performance, both stereotypes of gender and of politeness. My principal question underlying this study is: how can we develop a complex, pragmatic model of interaction which can account for the way that gender, in its interactions with other variables like race, class, age, sexual orientation, contextual elements, and so on, inflects the production and interpretation of linguistic politeness and impoliteness? Crucial to this project is a wider critique of many of the linguistic models available at present. My dissatisfaction with models of language production and interpretation developed within linguistics, and also with models of gender formulated within feminist theory, has led me to try to find new ways of approaching the analysis of politeness. Linguistic politeness lies implicitly at the heart of a great deal of gender and language research, from Lakoff (1975) onwards – the 1

2

Gender and Politeness

notion that women are more polite or deferent than men underlies the analysis of a range of linguistic features, from tag-questions to directives. My aim in this book is to foreground and challenge these stereotypical assumptions about gender, and to develop a new, more contextualised form of analysis, reflecting the complexity of both gender and politeness, and also the complex relation between them. Theorists such as Holmes (1995), who asserts that women are more positively polite than men, have tended to adopt a very functional form of analysis, whereby they argue that particular language items or strategies can be simply classified as polite. This enables such linguists to undertake quantitative research and measure whether women are more polite than men. However, this assumption that politeness can be so easily codified is one which I contest, since it is only participants in specific communities of practice who are competent to judge whether a language item or phrase is polite for them or not. Linguistic model In this study, I shall be drawing largely on critical work which has been undertaken in feminist linguistics, and which can be broadly termed Third Wave feminist linguistics, in order to question current gender and language research which tends to remain at the utterance level of analysis and focuses on the language production of individual speakers (Christie, 2000; Cameron, 1998a; Mills, forthcoming, b; Bergvall, Bing, and Freed, 1996). Third-Wave feminist linguistics is a form of anti-essentialist analysis which is critical of Second Wave feminist linguists such as Dale Spender, Robin Lakoff, and Deborah Tannen for their focus on a homogeneous ‘women’s language’ which they assume is the result either of the oppression of women or of the different socialising of women and men (Spender, 1980; Lakoff, 1975; Tannen, 1991).1 The type of analysis which is developed within this book is more concerned with the analysis of the role of gender in language production and interpretation. It is concerned not with the individual utterance but with the coconstructed nature of talk within a particular context; therefore it is more influenced by work within pragmatics which focuses on the interaction of individuals and context (Sperber and Wilson, 1986; Christie, 2000). I have also drawn on critical work in integrational linguistics, such as Toolan (1996), which forces us to reassess our

Introduction

3

presuppositions in constructing models of language production and interpretation. Rather than assuming that males and females speak differently, I have critically modified Wenger’s (1998) term ‘community of practice’, which is concerned with analysing groups of people who are drawn together in the performance of a particular task. This notion of community of practice has been particularly influential in feminist linguistics, as can be seen in the articles in the special issue on the notion of community of practice and gender in Language in Society (Eckert and McConnell-Ginet, 1998, and 1999). This work aims to produce a more context-based model of gender, where gender construction is constrained by its negotiations with suppositions of community rules of appropriacy and stereotypes. I argue that Eckert and McConnell’s modified notion of community of practice is insufficient to describe the complex negotiations which take place between individual speakers and the various linguistic communities of which they are members, simply because that model views that interaction as governed largely by constraint (Eckert and McConnell, 1998, 1999). In this book I map out the ways in which individuals negotiate with what they assume are community-of-practice norms for linguistic behaviour. I draw on a model of the relation between speakers and their communities which is more concerned with the discourse level (that is the level of structures above the utterance). A concern with analysis at the discourse level is one common to many discourse analysts. However, because discourse is often used in very vague ways by theorists, it is important to be clear what I mean by discourse level here. My model of discourse is one informed by the work of Michel Foucault in that I regard the production and interpretation of language as rule governed – governed at a level which is not within the control of individual speakers or indeed of institutional forces (Mills, 1997, and forthcoming, a). However, because this model often presents a view of the individual speaker as simply the intersection or the effect of discursive forces, I have tried to reinscribe the role of individuals, without falling prey to the liberal humanism which often accompanies concerns with the individual speaker. I am concerned with the negotiation that takes place between individual speakers and their communities of practice and the wider society, which Bourdieu’s (1991) notion of ‘habitus’, particularly as modified by Eelen (2001), seems to capture. I argue that there is a productive conflict within communities of practice, and that

4

Gender and Politeness

communities of practice are not as hermetically sealed as Eckert and McConnell seem to suggest in their work (Eckert and McConnellGinet, 1999). The norms of one community often spill over into another because group members belong to many different linguistic communities and sub-groups, and members do not necessarily feel comfortable or welcome within particular communities of practice (Bucholtz, 1999b). Furthermore, community members do not necessarily agree on which forms of behaviour are the dominant ones in any particular community of practice. Since the boundaries of linguistic communities are not clearly demarcated, there is a certain fluidity in what is considered by each member to be appropriate. Communities of practice are in a constant process of change, determined by the actions and assessments of individual members in relation to the group. The notion of change brought about by individual members of the group in relation to perceptions of group norms is important in thinking about how individuals and groups assess politeness. As I discuss in chapter 2, appropriateness is a problematic concept but one which informs linguistic production and interpretation. At the level of individual speakers, the assessment of appropriateness is important, but within a community of practice there might be a wide variety of judgements around the notion of what is and is not appropriate. Eelen’s work is important in seeing the dynamic nature of the judgement of appropriacy in relation to linguistic acts, but what I should like to add to his work is the sense of the role of stereotype, and the force of this on our own negotiations with notions of appropriacy (Eelen, 2001). I also feel that it is important to analyse the force of the wider societal and institutional pressures on individuals and the negotiations which individuals engage in with forces beyond the level of the community of practice. It is this dynamic nature of communities of practice and the often conflictual relation of individuals to particular communities of practice which is central to my work and which significantly extends the concept of community of practice beyond its current usage. Gender In order to develop a model of gender which is adequate to the task of describing the complex negotiations which the gendered subject undertakes, in chapter 4 I fuse a model of performativity drawn

Introduction

5

from the work of Judith Butler (1993) with a model of gendered domains developed by Alice Freed (1996) and Bonnie McElhinny (1998): that is, a model of gender as an act, or as a verb, which is enacted within specific environments and contexts which are themselves gendered by association (Crawford, 1995). I bring these two theoretical positions together as it enables me to discuss the instability of the gendered identity constructed within language but also to analyse the way that gender is dispersed into contextual elements rather than being located at the level of the individual. This fusion enables me to describe gender at the level of discourse rather than only at an individual and utterance level, whereby settings, strategies, discursive moves are normatively gendered by communities of practice and negotiated, contested, affirmed, and, crucially, changed by community members. Thus, rather than the utopian vision of gender, sometimes suggested by the work of Butler (1990, 1993), as something which one can perform as one wishes, gender here is performed but within constraints established by communities of practice and our perceptions of what is appropriate within those communities of practice.2 Furthermore, in order to analyse gendered practices, I shall analyse the way that one’s sense of one’s own and other’s gender identity is already raced and classed and affected by other variables such as perceptions about education, age, sexual orientation, and the extent of mutual knowledge. I also question the assumption that women are powerless and that men are powerful, drawing on recent work which has questioned the way that power has been theorised (Thornborrow, 2002; Manke, 1997; Diamond, 1996). Thus, my aim is to question much of the research on gender and language and to formulate a new theoretical model of language production and interpretation and its relation to factors such as gender. Politeness Much research on linguistic politeness remains at the level of the utterance. As I argue in chapter 2, since Brown and Levinson’s (1978/1987) ground-breaking work on this subject, most theoretical work has remained caught up within their theoretical and analytical paradigms, and, whilst there have been some excursions into Relevance theory, little has been attempted which would make for a

6

Gender and Politeness

pragmatic, context-based analysis of linguistic politeness and impoliteness. For example, many theorists simply assume that politeness is a set of behaviours which can be interpreted unequivocally: ‘In everyday usage the term “politeness” describes behaviour which is somewhat formal and distancing, where the intention is not to intrude or impose . . . Being polite means expressing respect towards the person you are talking to and avoiding offending them . . . politeness [is] behaviour which actively expresses positive concern for others, as well as non-imposing distancing behaviour’ (Holmes, 1995: 5). Thus, Holmes assumes that it is possible to categorise unequivocally those utterances in which an individual speaker affiliates to others or distances her/himself from others through language. My argument is that, even for individual participants in a conversation, it might not be clear whether they or their interactants are being distant or showing concern in any simple way. Even where theorists are prepared to be slightly more flexible about their notion of what constitutes politeness or impoliteness, they nevertheless tend to assume that most people would normally consider an act to be clearly polite or impolite: thus, ‘any utterance which could be interpreted as making a demand or intruding on another person’s autonomy can be regarded as a potential face-threatening act. Even suggestions, advice and requests can be regarded as face-threatening acts, since they potentially impede the other person’s freedom of action’ (Holmes, 1995: 5). What I am trying to bring to the research on politeness is a questioning of the notion that most people would agree about what constitutes a polite or impolite act. Indeed, influenced by work on miscommunication and misunderstanding, it is my contention in chapter 3 that disagreement about politeness, and particularly about impoliteness, is one of the constituent and defining features of politeness (Coupland et al. (eds.), 1991; Grimshaw (ed.), 1990). In a recent BBC radio programme about politeness, even though all of the participants interviewed stressed that they thought that politeness was very important, particularly in interactions between strangers, there were conflicting views about what constituted polite behaviour (BBC, Radio Wales, 2001). Such is the sense among certain members of the community that politeness is under threat or that there are changing forms of appropriate behaviour, that a group called the Polite Society, established in 1996, set up a Campaign for Courtesy and has designated a particular day each year as a

Introduction

7

‘day of courtesy’ (Gregory, 2001a). Its founder member, the Revd Ian Gregory, states that he has ‘noted the steady decline in considerate behaviour at every level of life in the United Kingdom’ and suggests that there may be a link between this decline in courtesy and what he sees as a general decline in standards. He argues that ‘some of us are still appalled and angry at the amount of selfish and ignorant behaviour that bedevils a once great nation’ and he also claims that ‘millions yearn for better standards of inter-personal behaviour’ (Gregory, 2001a: 2). He characterises courtesy in three words ‘to LISTEN, to SMILE and take TIME in our dealings with each other’ (Gregory, 2001a: 2). Gregory argues that ‘We live in a world dominated by brilliant science and technology. And yet in terms of human behaviour and relationships we seem for much of the time to be Neanderthal’ (Gregory, 2001a: 3). This type of characterising of the perceptions of the current state of politeness as an indicator for an assumed current moral decline is taken up by many other social commentators in the media. Dalrymple argues in an article in The Spectator that the calling of patients in hospitals by their first names is a ‘sign of the ever-greater vulgarity and shallowness of British life’ (Dalrymple, in Gregory, 2001b: 2). Daley, in an article in The Daily Telegraph, argues that ‘there has been a fundamental change in the national view of civility: courtesy is now confused with deference. Over the last generation, the British have been taught to believe that politeness was just a bourgeois trick to keep social inferiors in their place. Is it too late now to say that equality does not amount to everyone learning to say “I can do anything I want”?’ (Daley, 2001: 3). In this type of thinking roadrage and air-rage have become emblematic of a shift from respect for community values to egocentrism and selfishness. In many of the newspaper articles and phone-ins which I surveyed there was a uniform feeling that politeness standards have declined. For example, in an article in The Times, Celia Brayford suggests that ‘life in Britain has become an unending struggle against negativity and nastiness. The courtesy for which the stereotyped English were once ridiculed has been transformed into habits of casual, everyday brutality which astonish our continental neighbours’ (Brayford, cited in Gregory, 2001c: 4). However, responses to Brayford’s article varied and many readers wrote in to state that they did not find that British people in general were any less polite than other nations, or than

8

Gender and Politeness

British people were at other periods of history (Gregory, 2001c). It is this sense that politeness is an issue that people feel that they need to debate and perhaps defend will be reflected in this book, rather than assuming that politeness is a fixed and easily recognisable linguistic phenomenon. Many theorists of politeness assume that they know what politeness is, and they make a clear distinction between ‘folklinguistic’ notions of politeness and linguistic politeness itself (Eelen, 2001). Watts et al. (1992) suggest that we should make a distinction between firstorder politeness, that is ‘the various ways in which polite behaviour is perceived and talked about by members of socio-cultural groups’, that is common-sense notions of politeness; and second-order politeness, that is the ‘theoretical construct, a term within a theory of social behaviour and language usage’ (Watts et al., 1992: 3). Eelen terms these ‘politenessl’ and ‘politeness2’, and sees one of the major theoretical problems in politeness theory to be the confusion between which one is being discussed in analyses; he suggests that politeness theorists ‘should avoid getting involved in the struggle over representations of reality, and instead incorporate these representations into reality by making struggle over them the object of research’ (Eelen, 2001: 46). I consider this distinction between folklinguistic/first-order politeness/politeness1 and theoretical linguistic definitions of politeness/second-order politeness/politeness2 to be less easy to maintain than Eelen, since politeness is always by its very nature a question of judgement and assessment. Thus, folklinguistic notions often colour the theorists’ categorisations of utterances. I would agree with Eelen that sometimes difficulties with analysis arise because theorists rely too much on their own personal assessment of whether an utterance is polite or not, whilst claiming that they are analysing objectively. In contrast, what I focus on throughout this book is the analysis of what people judge to be polite and this involves me both in discussing with individuals what they consider to be polite and also examining the way individuals relate to each other in conversations. I argue, throughout this book, that it is not a simple matter to identify those utterances which are non-imposing or which show concern for others, either as a participant in conversation or as an analyst. Rather than simply analysing decontextualised utterances, I shall attempt to develop a pragmatics-based model of interaction

Introduction

9

to account for the way that participants make sense of politeness and impoliteness, and the way that they vary their own production of politeness and impoliteness according to their assessment of the context and the particular community-of-practice norms. My principal contention is that politeness cannot be understood simply as a property of utterances, or even as a set of choices made solely by individuals, but rather as a set of practices or strategies which communities of practice develop, affirm, and contest, and which individuals within these communities engage with in order to come to an assessment of their own and others’ behaviour and position within the group. Thus, this type of analysis is not simply concerned with the intention of individuals to be polite or impolite, nor is it simply prepared to add the perlocutionary effect of their talk on others to this paradigm, so that the hearer is included in a rather perfunctory way. Instead, I shall be arguing for a fundamental rethinking of the way that we analyse intentions and interpretations of linguistic acts – the relation between speakers and hearers. As I mentioned earlier, I also draw attention to the fact that misunderstanding needs to be at the centre of our analytical model rather than the assumption being that there is perfect communication between interactants (Wodak, 1998; Coupland et al., 1991; Grimshaw, 1990). In all interaction, individuals are working out their gendered identity and their position within a community of practice, as well as communicating with others, and politeness and impoliteness play a key role in presenting and producing a particular type of identity, and negotiating a position in the community of practice. Judging someone’s utterance to be polite or impolite is also making an assessment of them as individuals. Deciding to be polite or impolite is a crucial part of constructing one’s own sense of identity as ‘nice’, ‘considerate’, ‘assertive’, or ‘tough’, and assessing one’s role in relation to other members of a group. Within all communities of practice there is conflict over meaning and over the notion of what is appropriate. What is appropriate linguistic behaviour is implicitly a key notion in all research on linguistic politeness, but it is one which is rarely described or analysed adequately. Factors of gender, race, class, age, education, and knowledge play a major role in assumptions about the level of appropriate linguistic behaviour within particular communities of

10

Gender and Politeness

practice. I analyse the norms of the particular community of practice at a higher level than simply the utterance, in terms of what forms of politeness/impoliteness are considered to be permitted to whom; what strategies may be adopted by participants and how they are judged by others; what contexts are implicitly or explicitly gendered in terms of what style of utterances or move is considered appropriate and for whom. I am not arguing that interactants always behave appropriately. In fact, it is precisely the contestation of these norms that people seem to be concerned with in everyday interaction. A concern with appropriateness and the way that people assess others’ utterances involves analysing the way communities of practice perform and monitor politeness and impoliteness over long stretches of talk, describing the function of individual acts or interactions over an extended period of time, and the effects of particular acts over time. Thus, I am concerned to develop a form of analysis which can describe the assessments made by individuals of their talk in relation to community judgements of appropriacy (which they themselves have helped to form). Necessarily, therefore, the definition of politeness which I use in this book is not a simple one and is not concerned to try to prove that certain participants in a conversation are more or less polite than others; rather, I aim to develop a form of analysis which focuses on the judgement of both one’s own speech and that of others in relation to notions of politeness and impoliteness and the functions which such judgements have within particular communities of practice. Methodology and data This book tries to analyse the intuitions that individuals have as speakers of a language who are trying to understand others, and work out their position in relation to other speakers, rather than trying to formulate a grand universalising theory which may have little to do with language as it is experienced in particular contexts. There seems to me to be a number of serious problems with data collection and interpretation in general, which I discuss more fully in chapter 1. Whilst I feel that it is essential to draw on ‘real’ data (audio-recorded conversations) in conjunction with other kinds of information about language, linguists need to be very wary of how data is analysed. When linguists interpret data, they often use the

Introduction

11

type of outdated techniques which in literary criticism would be termed New Criticism: a type of close reading of a text which only attends to the formal textual elements and does not consider questions of interpretation or context. This is particularly problematic in the analysis of conversation, since each participant will have a slightly different take on what is going on in the interaction. Counting the length of utterance, the number of modal verbs, or the number of questions each participant asks might be useful in certain types of linguistic analysis, but in the analysis of politeness it can tell us little. This type of analysis in linguistics often attends only to the formal features and therefore does not see the effect of the utterance on, say, the hearer’s self-perception or the motivation of the speaker in making what could be termed a formally polite utterance.3 This type of formalist analysis is clearly not adequate as it does not begin to describe the complexity of the use and interpretation of politeness. Conversations are, among many other things, the instantiations of our emotional responses to others. We decide whether we like certain people, whether we can trust them or not, on the basis and in the course of what they say and the way they say it. Politeness as a process within language is very closely tied to this process of working out our emotional responses and relations to others, but this level of behaviour is rarely considered within linguistic accounts (for an exception, see Wierzbicka, 1999).4 A further factor which is often ignored in linguistic accounts of politeness is that interactants have a number of signifying practices, of which we are largely unaware, which we use both to signal to others our overall intentions and to try and interpret what others’ overall intent is. When we make an overall assessment of what is happening in a particular discussion, we as interactants may well latch on to one or two phrases which someone said in order to characterise what someone was really saying. This sense of constructing the gist of a conversation is a constant process of hypothesising, but it is very important in terms of thinking about conversations in the past in relation to present conversations. Formalist analysis, because it is wholly synchronic, is unable to see the importance of this past history or this more globalising way of considering how people understand utterances. It is also unable to analyse the way that people make guesses at what is going on in a conversation. Work on topic control, footings, and frames is of particular interest here and will

12

Gender and Politeness

be discussed in grater detail in chapter 1 (Goodwin and Goodwin, 1990; Schiffrin, 1990). I analyse various types of data within this book: audio-recordings of conversations, questionnaires, interviews, and anecdotes. However, this will not be a very conventional type of analysis. Many analysts state unequivocally what they take to be going on in a conversation. I feel that this is not possible. What strikes me as selfevident in a recording of a conversation was not necessarily salient for the interactants I have recorded; nor do I feel that it is possible to capture the complexity of what is going in a conversation, even when I have been present, or when I have conducted interviews with the participants after the recording. It is for this reason that I discuss the recorded data alongside a number of anecdotes, which others have discussed with me; this is not because I necessarily assume that these anecdotes are ‘true’ or more authentic than the recorded data, but simply that anecdotes give us an insight into the role of stereotype and show us the ways in which people make assessments of politeness, which I argue is crucial for an analysis of politeness.5 I also draw on data which I have collected from structured interviews, because it is important to find out what people think about politeness/impoliteness in general.6 In analysing this recorded data, interviews, and anecdotes, instead of making claims that, for example, women are more polite than men, I shall not be generalising from this data to any great degree. I do not believe that it is possible to generalise from this very small sample to claim that everyone has the same beliefs about politeness or that everyone performs in the same way as the individuals whom I have recorded – I shall rather be using this body of data to illustrate key points in my argument about politeness and impoliteness, and to question the presuppositions of many politeness theorists, in order to suggest that a different form of analysis from the conventional Brown and Levinson model is needed. In this sense, this study is primarily theory-driven, because it is at this level that most change has to be made in the analysis of politeness, as Eelen has suggested (Eelen, 2001). Analytically, much of the burgeoning research on politeness is of interest, but because it is based on Brown and Levinson’s work in the main, the results of such research are problematic. This theory-driven approach however is underpinned by extensive data collection and analysis and by

Introduction

13

trying to confront the problems which data-collection and analysis pose for linguists. Much of the data is analysed in a qualitative way, since a focus on communities of practice necessitates this kind of detailed and careful analysis (Wodak, 1998). Structure of the book In chapter 1, I consider the general problems with linguistic interpretation, most notably the model of the speaker, the hearer, and communication. I then go on in chapter 2 to consider the problems with Brown and Levinson’s work on politeness. I try to draw together the various critiques that there have been of Brown and Levinson’s work, so that new forms of analysis can be considered. In chapter 3, I analyse the way that politeness has often been discussed in relation to impoliteness, but I argue that impoliteness has rarely been analysed in its own terms. In this chapter I focus on how interactants come to decide that someone has been impolite to them. In chapter 4, I examine more process-oriented and performative theorisations of gender, in order to examine, in chapter 5, the stereotypes which abound in research on language and gender. Whilst notions of stereotype are crucial for interactants in order to come to an assessment of appropriacy, I argue that these are hypothesised stereotypes which differ from individual to individual. Thus, stereotypes have an effect on interpreting and producing speech. However, I argue that analysts often unproblematically assume that their own hypothesisation of gender stereotypes is generally applicable and shared by others; they often draw on these stereotypes when analysing gender, for example, assuming that all women are more polite than all men. However, whilst interactants might draw on a version of this stereotype in their interactions in certain contexts, that is not to say that it is an accurate analysis of their linguistic behaviour as a whole. Thus, we need to make a clear distinction between the hypothesised stereotypes drawn upon by interactants and indeed challenged and ridiculed by some of them, and our own theoretical view of gender which informs the type of model we use to analyse and the research questions that we ask. In the Conclusions, I discuss the implications of this type of work for future work in gender and language research and in politeness research.

14

Gender and Politeness

To conclude, my main contention in this book is that theorists of linguistic politeness need to reorient their work so that they do not make false assumptions about what is going on in conversation when people judge each other as being polite or impolite. What we need are new ways of analysing politeness so that we can see the varying forces at work in the process of being polite and impolite, and the outcome and effects of these assessments. I argue that we should not focus on, for example, the analysis of indirectness as an instance of polite behaviour, but rather that we should ask fundamental questions about whether all of the participants in the conversation we are analysing consider particular utterances as indirect and whether they themselves consider indirectness to be indicative of politeness or not. Since politeness and impoliteness are matters of assessment, unless we focus on the process of judgement itself, we shall present only our own personal assessments of politeness, rather than a theorised analysis. Notes 1. It should not be assumed that Third-Wave feminist linguistics is an agreed-upon term within gender and language research. It is adopted here because it has the advantage of questioning early feminist assumptions about gender and language but at the same time still locating itself within feminism, which terms such as postfeminism do not. 2. Butler herself does not necessarily see the performative in this simplistic way, as I argue in ch. 4, but she has been interpreted as making these assumptions by others. 3. I shall use the term ‘formal politeness’ to refer to those ritualised phrases such as ‘please’ and ‘thank you’ which most people recognise as polite. I shall be questioning later in this book whether these elements are always interpreted or indeed are always intended to function as indicators of politeness. 4. Whilst Wierzbicka’s work is an exception, as she does consider the way that emotions are represented in language, she unfortunately often assumes that generalisations can be made cross-culturally about universals in the expression of emotions, which I would contest. 5. The data consist of over 66 hours of audio-taped recordings which I have supplemented with a select number of interviews with participants. These case studies range from single-sex and mixed-sex informal group discussions to single- and mixed-sex formal meetings within a range of private and work/leisure settings. The settings include: departmental meetings, office interaction, dinner parties, reading groups, student supervisions, political party meetings, parish council meetings, and interactions in

Introduction

15

primary schools. More details of these data are given in chapters 3 and 5. I have drawn on these data selectively to illustrate points made throughout the book. 6. I interviewed 20 adults from a range of different occupations and age ranges, and 4 children about politeness, using a set of standard questions about politeness. The questions asked them to consider what type of event they considered polite or impolite, and asked them to talk about an event which they had considered impolite, polite, or overly polite.

1

Rethinking linguistic interpretation

Introduction This chapter is concerned with general problems with linguistic interpretation. This concern has arisen because of the problematic models often used by linguists implicitly when they discuss speakers and hearers and their agency, and the problems associated with the way data are collected, analysed, and interpreted in linguistic research in general. Because of my interdisciplinary work in literary theory and post-colonial theory, where many Cultural Analysts scrutinise carefully the foundational elements in their research work, I have found difficulty with the common-sense models used in linguistics when approaching the analysis of data. Literary and cultural theory have, since the 1970s, questioned issues such as intentionality, interpretation, agency, and the role of the critic/analyst, and have become more cautious or sceptical than many linguists in making claims about the status of their analysis (Kendall and Wickham, 1999). Many linguists, however, have made problematic claims for the status of their own work and the generalisability of their findings. Many of the problems that I will be discussing in chapter 2, in analysing politeness, are problems encountered by linguistic research in general, and therefore I deal with these general problems in this chapter before going on to assess the more specific problems faced in linguistic models of politeness. Here, I examine the types of assumptions which are implicit in a range of different linguistic subdisciplines (for example: Discourse Analysis, Critical Linguistics and Critical Discourse Analysis, sociolinguistics, Conversation Analysis, pragmatics, social psychology, and ethnography), in order critically to examine the models and methodologies which they use and the 16

Rethinking linguistic interpretation

17

difficulties which I feel are entailed in their use.1 I would like to situate myself outside of the conflicts between Conversation Analysis (CA) and Critical Discourse Analysis (CDA) which have raged in recent times, and which have resulted in quite entrenched positions with regard to analysis (see for example, Schegloff, 1997).2 Rather, I want to develop a form of analysis which combines the most productive elements of both forms of theorising, combining a clear political position with a careful and systematic analysis of the determinants of discourse (see for similar attempts, Thornborrow, 2002; Shaw, 2002).3 Thus, the critique in this chapter is not criticism for its own sake, but is a crucial stage in developing more adequate models for the analysis of language use. I shall discuss the problematic aspects of linguistic analysis at the following levels; the speaker, the individual’s relation to the group, the model of communication and language, and methodology. The Model Speaker Perhaps one of the most problematic aspects of linguistic research in general is the model of the speaker and hearer which is drawn on implicitly and sometimes explicitly. Generally, linguistics assumes that the individual can be discussed unproblematically as an autonomous person, who chooses to use certain language items and strategies rather than others. Whilst there are obvious methodological reasons for choosing to focus on this type of Model Speaker who is in control and who is not subject to moods, memory loss, or seemingly irrational behaviour, since it is easier to predict the behaviour of such an individual and to make generalisations, this notion of the autonomous Model Speaker presents us with a number of problems. What is not considered is the historicity of the model of the individual itself, the fact that it developed at a particular historical moment in response to philosophical and socio-economic circumstances. Scollon and Scollon trace the history of the development of the model of the atomistic individual within Western cultures, and show how it developed at a particular historical conjuncture, primarily from Utilitarianism, but its development was the result of a range of different forces – philosophical, political, and economic (Scollon and Scollon, 1995). This model of the individual has influenced the use of the Model Speaker in much linguistic work, and as the

18

Gender and Politeness

Scollons argue ‘this self is understood as being opposed to what are referred to as “traditional social roles”’ (Scollon and Scollon, 1995: 208; see also, Werkhofer, 1992). Thus, this autonomous speaker is very much a product of a broadly colonial stereotypical binary opposition of those in the West who are ‘free’ and who determine the course of their own lives, and those whose selves are conditioned by societal roles and who are thus deemed incapable of independent thought and action (Azim, 1993). In other fields of research, for example literary and cultural theory, the individual is not generally assumed to be a self-contained unit, but rather a process, a site in construction through interaction. For example, within psychoanalytical theory, the self is not considered to be an isolated Descartian individual who is in full control of what s/he says, but rather an unstable and rather fluid ‘holding operation’, where unconscious and repressed wishes and desires are held at bay, negotiated with by an imperilled ego (Kristeva, 1982, 1984). It is obviously difficult in linguistic research to draw on this more complex model of the self where flux, uncertainty, and change are more salient than an assumed stability and control, but it seems to me, in pragmatics research in particular, this notion of the self must be integrated more, or at least there must be an acknowledgement that unconscious and possibly irrational motivations play a role in language production and reception. When speakers produce utterances, it is not just their conscious minds which construct utterances, but their unconscious motivations also play an important role. Whilst it may not be possible to access and document those motivations in any systematic or convincing way, we need nevertheless to move away from the notion that we, as analysts, know exactly what motivated the speaker or that the speaker knows and is conscious of all of his/her motivations for speaking in a particular way.4 If we take the example of the way that certain individuals relate to authority, we can see differences of a general nature which affect an individual’s overall linguistic performance; thus, for a subject who has great difficulty relating to authority figures, this reaction might play a major role in the way that they formulate polite utterances and may help to account for certain types of defensive behaviour or over-reactions to such strategies which they interpret as interruption or direct commands. That is not to say that a linguist should try to discover whether the subjects they are investigating have difficulties in relation to authority,

Rethinking linguistic interpretation

19

but to suggest that a person’s overall linguistic performance may be affected by factors which are beyond the individual’s conscious control, and which may perhaps always lie outside the analytical framework of the linguist. I find Foucault’s conception of the self, or rather the death of the self, where the self is only an arena where certain discursive elements are brought into play, much more challenging and theoretically convincing (Foucault, 1978). However, even within this antipsychoanalytical model which seems to be trying to eliminate the subject or the self from analysis, it is still possible, when analysing the self as an effect of discourse, to consider unconscious motivations as part of the overall analysis (Skeggs, 1997; Liladhar, 2001). It is this awareness of the role that unconscious motivations play in the construction of talk that I would like to stress, along with the awareness that the controlling ego which linguists describe as the individual is only a small part of the self or subject. The importance of focusing on the whole individual, or at least being aware of other forces than the ego at work in the individual’s linguistic production, forces us to challenge the notion that speakers and analysts know the full extent of what is going on in an interaction. I should be clear that I am not saying that speakers do not act from choice and produce utterances that they have formulated within their conscious minds, as I shall show later in this chapter. However, speakers do not necessarily decide consciously to utter every word that they say; there are other factors, such as subconscious motivations, verbal routines, and social pressures which play a major role too. Literary and cultural theory, particularly since poststructuralism, rather than viewing the language production of individuals as a product, have seen language as the site where identity is constructed. Language has assumed an almost mythic role in the analysis of the process of the construction of the self, and indeed language or discourse for many theorists is a major determining factor in subjectivity. For many psychoanalysts, identity construction is something which is never achieved, but always in process. For some linguists, this view of subject construction has also been important; for example, for many ethnomethodologists, and social psychologists using discourse analysis, the process of construction of identity is precisely what is focused on in analysis, or indeed the way that identity is brought into play in conversation. Antaki and Widdicombe, for

20

Gender and Politeness

example, state that they do not want ‘to treat people as informants, nor do they want to interpret what people say, still less speculate on the hidden forces that make them say it. Rather, they want to see how identity is something that is used in talk, something that is part and parcel of the routines of everyday life, brought off in the fine detail of everyday interaction’ (Antaki and Widdicombe, 1998: 1). Rather than seeing identity as separate from language, they argue that ‘membership of a category is ascribed . . . avowed . . . displayed in local places and at certain times, and it does these things as part of the interactional work that constitutes people’s lives . . . not that people unilaterally have this or that identity which then causes feelings and actions, but that they work up and work to this or that identity’ (Antaki and Widdicombe, 1998: 2).5 Schegloff argues that it is the categories which participants orient to which must be seen as ‘privileged in the constitution of socio-interactional reality’ and which ‘therefore have a prima facie claim to being privileged in efforts to understand’ that reality (Schegloff, 1997: 167). However, I do not wish to assume that identity is solely constructed through interaction, as there are many other processes at work in interaction (such as the imparting of information, the achievement of actions, the resolution of problems, and so on) which may play a role in the construction of identity but which also achieve other goals. Furthermore, identity is shaped and worked on by other wider forces outside interaction, such as class, race, education, access to resources, and so on, and whilst I would agree with Gumperz that ‘an individual’s choice of speech style has symbolic value and interpretive consequences that cannot be explained simply by correlating the incidence of linguistic variants with independently determined social and contextual categories’ (Gumperz, 1982: vii), I nevertheless feel it is important to be aware that factors such as race, class, and gender do affect the way that we speak and interpret. One of the aims of this book is to try to examine both the construction of a gendered, raced, and classed self as a process within the interaction, but also as a process which takes place within social constraints profoundly affecting the form of the interaction and the self. Many linguistic models analyse utterances as if the speakers choose each word individually; given the rate at which speakers generally compose utterances, this is obviously an overly strategic view of the process of producing speech. Many Conversation Analysts

Rethinking linguistic interpretation

21

assume that each response, hesitation, eye movement, gesture has significance and is chosen for strategic purposes (Goodwin, 1981). I would question this assumption; it is necessary to understand how it is that speakers manage to produce utterances without having to choose each particular word. Within the language as a whole, and within certain contexts, there are pre-fabricated phrases and clauses, set responses to certain cues, and linguistic contexts in which there are limited choices to be made (Mills, 1995a). This is not just a case of adjacency pairs and formulaic utterances, but also stretches of speech which are relatively predictable and which one would expect to occur in certain contexts (Atkinson and Heritage, 1984). With the frame of context developing as the conversation develops, it is possible to limit the range of interpretations which can be given to certain utterances and also to the potential choices about what to say. Rather than seeing the individual utterance as something which is produced by the speaker at that particular moment, invented anew, as it were, we also need to see the utterance as a result of a longer process of thinking, habit, and past experiences. Someone’s interactional history helps to set up certain habits which predispose them to use language in particular, though not entirely predictable, ways. As Toolan remarks: ‘while language is never a code, it is apparent that individuals become habituated to a code-like predictability of usage, forms and meaning. Nor need the habituation be thought of as necessarily undesirable or demeaning; it will often be a constructive acceptance of the limits of communicative innovation and desirability of recognisably shared public forms’ (Toolan, 1996: 9). What Toolan is arguing for is a highly individualised sense of each interaction and the type of work and processing that each individual undergoes in the process of interacting with others; whilst I would wish to focus more on the group and social level of these scripts and situations, individual motivation and sense-making springs from the interaction of a range of factors which cannot be encapsulated within a simple notion of the autonomous human subject making choices about what to say.6 Toolan goes on to argue that, in time, our relation to our own history becomes more schematic and more important in terms of interpreting utterances: ‘Part of the human response to finiteness and normativity is the tireless schematising that we evidently undertake, the sorting of past experiences into remembered scripts, activities and stereotyped situations. It is

22

Gender and Politeness

through this shifting multi-dimensional mental network of scripts, situations and styles that we understand the making of contextualised sense of particular episodes of linguistic interaction’ (Toolan, 1996: 9). Our thinking about what is possible for us to say is predicated on what we see as our interactional history within groups and with particular speakers; but that involves a certain selectiveness about what we think that interactional history consists of. Individuals over time develop certain linguistic habits, predispositions to act within relatively predictable ways with language. Thus, we notice these regularities in others and we assess that person’s linguistic behaviour in relation to these personal norms. For example, some couples develop particular styles with each other which may appear to others quite aggressive and combative. For some, it may take a while to become used to this as playful interaction rather than as argument, and indeed it may never be accepted by others as a form of verbal play. But their assessment of the particular interactions of this couple will be based on their knowledge of their habitual style of interaction; thus, what might be judged to be aggressive behaviour in others might not be in their case. Individuals expect others to judge them on the basis of their verbal habits; particularly in relation to politeness, they expect to be judged not only in relation to the community-of-practice norms or the wider society’s norms, but rather in relation to their own particular habitual style, as I show in chapters 4 and 5. A further problem with the focus on the individual is that analysts often focus on the intentions of the speaker alone. Within literary theory, intentionality has been much debated. Whilst traditional critics often felt that it was their task to retrace the intentions of the author in the text, in order to discuss the meaning of the text, poststructuralist theorists have asserted that such retracing is always a post-hoc rationalisation by the critic. For many literary theorists, the meanings available in the text will always exceed the intentions of the speaker/writer (even if we could manage to trace those intentions in any simple way) (Mills, 1994a). In the analysis of utterances, this question of intentions is crucial, for one’s motivations and intentions are often formulated in the process of interacting with someone else, and may not become clear to either party until after the interaction has taken place. There is a certain amount of self-justification which takes place when someone asks

Rethinking linguistic interpretation

23

us about our intentions in an utterance: we generally put our intentions retrospectively in a positive light. Very often our motivations in interactions are very mixed – there may be several possible intentions at play during a conversation; only when the interaction has finished and we can make a decision as to what the interaction was globally about, do we decide what our intentions actually were. Furthermore, one intention might be primary in our minds when we utter the statement but, when we recast the interaction later on, we might reformulate our intention in the light of what happened later. That is not dishonest, but simply a part of the way that we reflect on our actions and utterances. Thus, our intentions and those of others cannot be seen as entirely achieved and thus cannot be assumed to be easily recoverable by analysts.7 Rather than dispensing with the notion of intentionality altogether, as much literary criticism has, Toolan argues that we should interrogate the notion of intention and use it in a more considered way, since hearers actively engage in the construction of a hypothesised intention for the speaker in order to make sense of utterances, even though, in the final analysis, intention itself is not recoverable: ‘intentionality, I argue, must in essence be the intentions that a hearer attributes to a speaker, without hope, possibility, or need of confirmation (by the speaker) of their accuracy’ (Toolan, 1996: 19). Thus, intention should be seen as a working hypothesis which each speaker and hearer formulates and revises in the light of subsequent utterances. Toolan argues that the notion of intentionality is indispensable, since it is an organising principle in the sense-making process in conversation; however, we cannot assume that we can accurately discover our own or others’ intentions. Judith Butler’s work on racist and sexist language has interesting implications here in relation to intentionality (Butler, 1997). She argues that many examples of ‘fighting words’ have the force to wound not because they originate with the individual but because some utterances have a more authoritative force because of their origin in, association with, or adoption by, institutions.8 Thus, a sexist statement may be that which is considered by the hearer to be intended to offend, but it is also a statement which seems validated in some way by its origin outside the individual speaker, and perhaps also its affirmation by institutions and by historical events. Butler states that rather than seeing such acts as discrete speech acts, we should

24

Gender and Politeness

see them as a ‘ritual chain of resignifications whose origin and end remain unfixed and unfixable. In this sense, an “act” is not a momentary happening, but a certain nexus of temporal horizons, the condensation of an iterability that exceeds the moment it occasions’ (Butler, 1997: 14).9 This notion is important not just in relation to certain types of utterances, such as racist and sexist statements, but also in relation to utterances in general. As mentioned above, linguists often analyse isolated utterances, as if they were simply chosen by the speaker at that particular moment without reference to previous utterances and contexts; the institutional history of certain types of utterances and phrases weighs on the production and reception of utterances.10 Whilst I am not suggesting that individuals have no power to create their own meanings in utterances, it is important to distinguish between those utterances which can be seen to be relatively creative and those which are more recycled or which seem to be determined by agencies outside the individual her/himself. We need to be aware of utterances having a history outside the individual. In terms of the conscious choice of the individual about the utterance that is produced, there are types of utterances for which the speaker has to assume responsibility; thus, racist and sexist statements are ones for which the speaker must be held accountable, even though those utterances are not necessarily ones which the speaker invented or created her/himself. In this sense, speakers have a choice over what they say and how they express themselves, but this notion of choice needs to be mitigated. This model of the self is not the same as the Model Speaker having complete control over all of the elements in their utterances, since, in some senses, the speaker is only accorded responsibility for what they say, rather than complete control and originality. Although, as Chomsky argues, individuals are potentially infinitely creative in language, this must be held against the Foucauldian notion that, in the face of this potential creativity, speakers, in fact, only invent within very tightly defined parameters, what Foucault describes as the ‘rarefaction’ of discourse (Chomsky, 1988; Foucault, 1981). In The Archaeology of Knowledge, Foucault tries to describe the ‘rules’ of discourse, which determine the production of utterances and texts, and, for him, these rules are specific to discourse, which changes according to certain autonomous discursive procedures (Foucault, 1994/1972). The individual has the

Rethinking linguistic interpretation

25

illusion of choice, but discourses themselves set out the parameters within which those limited choices can be made (Mills, 1997). Foucault makes a distinction between different types of discourse, or structures and rules for language production. There are those which have a more permanent nature and which play a role in producing other utterances, which act as a format for other speech acts, and there are those which do not endure. He states: ‘we may suspect that there is in all societies, with great consistency, a kind of gradation among discourses: those which are said in the ordinary course of days and exchanges, and which vanish as soon as they are pronounced; and those which give rise to a certain number of new speech acts which take them up, transform them or speak of them, in short, those discourses which, over and above their formulation, are said indefinitely, and are to be said again’ (Foucault, 1981: 57). Thus, for Foucault, the rules of discourse and discursive structures have a determining impact on the production of utterances, rather than the individual simply creating utterances in a vacuum. For many linguists, implicitly, the self of the speaker is seen to be stable over interactions and a person’s linguistic behaviour can be generalised from one particular conversation. However, it is clear from what I have said above that identity is constituted in conversation and changes over time, from context to context (Baker, forthcoming). If we consider cognitive anthropological work on the functioning of the brain and hence the functioning of language: ‘it makes no sense to speak of brains as though they manufactured thoughts the way factories make cars. The difference is that brains use processes that change themselves – and this means that we cannot separate such processes from the products they produce. In particular, brains make memories, which change the way we’ll subsequently think. The principal activities of brains are making changes in themselves’ (Minsky, 1986, cited in Foley, 1997: 9). This notion of the constantly changing individual needs to be built into our model of analysis. The notion of the Model Speaker must, therefore, be thoroughly scrutinised and the theoretical validity of making general assumptions about language use on the basis of this hypothesised ‘individual’ must be brought into question. When linguists analyse speech, they often focus on the dyad, the familiar stereotypical male and female couple, who formulate utterances on the following lines:

26 A: B:

Gender and Politeness ‘What time is it? ‘The post’s just been.’

This focus on two speakers makes analysis simpler, but it cannot be justified on theoretical or methodological grounds. Although a great deal of conversation does, in fact, occur between two speakers, for a variety of reasons, most notably ease of communication which is reinforced by technological constraints (i.e. e-mail, telephone, and so on, which are generally only available to two participants at a time), there are many situations where individuals, particularly in the work environment and in leisure time, interact as members of groups. It is obviously more difficult to analyse groups of people, largely because of the problem of clarity of recordings, but this technological problem should not force us to concentrate on the dyad, as if this were, in some ways, an exemplar of all other types of conversation. Even in the dyad, group values and community practices and strategies infiltrate and influence the individuals. Speakers and their utterances should not be analysed in isolation but in relation to a wider group or society. For many cultural critics, the relation to an Other is of primary importance, in terms of the construction of the self, but also in relation to the production of utterances. For Althusser, interpellation constitutes our sense of self; this notion of the ‘hailing’ power of the call of the Other to us forces us to recognise ourselves as a particular type of individual or within a particular role (Althusser, 1984). Thus, in the very act of constituting ourselves as a seemingly autonomous individual, we are constituted from the views, opinions and language of others. Butler defines this Other-orientedness as: ‘the primary dependency that any speaking being has by virtue of the interpellative or constitutive address of the Other’ (Butler, 1997: 5–6). Whilst I would not agree with this strong version of the Althusserian model of interpellation where one is called into existence by the address of the Other, nevertheless, the focus on a relation to an Other or, perhaps more productively, Others, as constitutive of the self is important. This Other-orientedness is perhaps the single most important element in language production. In Volosinov’s work, what we say is very much dependent on what we think the other person is likely to accept or understand, or what we think others expect from us or would like to hear; this notion of the other person is a hypothesis

Rethinking linguistic interpretation

27

on our part and is not the same as the hearer’s real wishes or views of the speaker (Volosinov, 1973). Thus, even in the formulation of what seem to be our own intentions, the Other plays a determining role. For Volosinov/Bakhtin, the individual takes in society’s values and mores through accepting, questioning, and modifying certain beliefs and ideologies, and by making them their own. Thus, for Volosinov it is difficult to make a clear dividing line between the individual and society, since the individual is created out of the elements of which society as a whole is constituted (Volosinov, 1973; Pearce, 1994). In his work challenging Western linguistic and psychoanalytical models of the self, Volosinov suggests that the self is constructed through a complex process of interaction between the social and the self rather than the self constructing itself in isolation, which seems to be the underlying assumption of most Western models of individuality. He sees the individual being constructed through an unconscious and conscious set of choices being made by the self in terms of the available resources and ideologies circulating in a particular culture or group. This relational view accords with many non-Western views of the self, where rather than being an autonomous individual, the individual is given a role or status only through its interaction with the larger group. For many cultures, such as in many Arab cultures, and in China and Japan, an isolated individual is viewed rather negatively. One’s sense of self and value is drawn from one’s position in relation to others and one’s role within the group and wider society as a whole. As Scollon and Scollon put it, in their analysis of East Asian ‘collectivist’ discourse styles: ‘individual members of a culture are not perceived as independently acting individuals, but rather they are seen as acting within hierarchies of kinship and other relationships. A son or daughter’s primary motivation for action, for example trying to get promotion, is thought to be to bring credit to his parents and to provide security for his own descendants. He or she is not thought of as acting on their own behalf or with their own interests and advantage in mind. Indeed, such individual action is seen as an aberrant or possibly pathological form’ (Scollon and Scollon, 1995: 131). I recognise the problem of generalising about other cultures, in the way that the Scollons do, and this view of other cultures is necessarily stereotypical as I shall describe later. However, I am trying simply to isolate the historical development for certain groups

28

Gender and Politeness

in the West of this type of individual-focused thinking. This model of the speaker used by linguists should be seen as Eurocentric, since it ignores other models of the self which hold within other cultures, and this creates difficulties when generalisations are made about other languages, as they are in politeness research (see chapter 2). I would argue that conversation and meaning is always coconstructed by all of the participants in a conversation and therefore the focus only on the speaker cannot be justified. Goodwin argues that interactants participate in engagement displays through the use of eye movement, gaze, and bodily posture, and this type of display mirrors the complex way that meanings are co-constructed: ‘The engagement display of one party shows an orientation to the displayed engagement of the other. However, that other party is also performing a similar analysis with the effect that his [her] display is simultaneously being organised with reference to the engagement state of the first. Each party’s body thus displays an analysis of what the other is doing and by that very display constrains what the other can and should be doing’ (Goodwin, 1981: 96). Some linguists, trying to find a way to explain the way in which interactants make sense of their conversation, analyse those elements which the speaker and hearer share. For some, the concept of background knowledge or members’ resources is used to describe this ‘shared’ knowledge (Fairclough, 1992). Speaker and hearer share certain types of encyclopaedic knowledge, which everyone can take for granted (for example, that certain trees lose their leaves in winter), and they can count on certain specific types of knowledge being taken as read in a particular interaction (for example, that the speaker has certain friends in common with the hearer and shares certain information about them). Sperber and Wilson have modified this model of shared knowledge which speakers draw on in conversations, since it is difficult for analysts to try to describe the range and, perhaps more importantly, the limits of this knowledge, and to indicate what elements of shared knowledge are salient at particular moments (Sperber and Wilson, 1986; Werth, 1999).11 Thus, rather than a body of knowledge which can be assumed to precede the conversation and which speaker and hearer draw on to make sense of each other’s utterances, Sperber and Wilson offer a model of mutual manifestness, that is those elements which, within the course of the conversation, are capable of being represented (Sperber and Wilson, 1986). This

Rethinking linguistic interpretation

29

is a marked improvement on the notion of background knowledge, since it is not necessary for the assumptions which are made to be true or accurate but simply representable or perceptible by speaker and hearer, in their particular cognitive environment. Sperber and Wilson give an example of Mary and Peter, that perfect speaker/ hearer couple, looking at a landscape where Mary has noticed a church; she says ‘I’ve been inside that church.’ Sperber and Wilson comment: She does not stop to ask herself whether he has noticed the building, and whether he assumes she has noticed, and assumes she has noticed he has noticed, and so on, or whether he has assumed it is a church, and assumes that she assumes it is, and so on. All she needs is reasonable confidence that he will be able to identify the building as a church when required to: in other words, that a certain assumption will be manifest in his cognitive environment at the right time. He need not have accessed this assumption before she spoke. In fact, until she spoke he might have thought the building was a castle: it might only be on the strength of her utterance that it becomes manifest to him that the building is a church. (Sperber and Wilson, 1986: 44)

This more complex notion of mutual manifestness is productive. However, either of these forms of shared knowledge, whether predating the conversation, as in mutual knowledge, or constructed in the interaction itself and indicated to the other in the conversation, as in mutual manifestness, assume that conversation is composed of elements which are taken as given within the conversation. I would like to focus on a more integrational and context-focused model which is concerned with sense-making and working things out, and perhaps also with getting things wrong, misunderstanding what is said, and tolerating a fairly high degree of misunderstanding. In sum, what the model of the ideal or Model Speaker, generally drawn on implicitly in linguistics, lacks is a sense of the difficulty of making sense and the indeterminacy of the meaning of utterances. The individual and the group Some linguists have started to move in the direction of analysing groups of speakers or the relation between groups of speakers and the individual, and how individuals define themselves linguistically in relation to groups (Eckert, 2000). Rather than using the notion of

30

Gender and Politeness

a linguistic community or speech community which seems to assume a homogeneous group of speakers, as I mentioned in the Introduction, I draw on the notion of the community of practice (Eckert and McConnell-Ginet, 1999; Holmes and Meyerhoff, 1999; Bucholtz, 1999b; Wenger, 1998). Wenger states that a community of practice consists of a loosely defined group of people who are mutually engaged on a particular task and who have ‘a shared repertoire of negotiable resources accumulated over time’ (Wenger, 1998: 76). Thus, the group defines itself and is defined by its set of linguistic practices which change over time and which are subject to negotiation by members. Eckert and McConnell-Ginet remark on the interactive nature of communities of practice, since ‘the development of shared practices emerges as the participants make meaning of their joint enterprise, and of themselves in relation to this enterprise’ (Eckert and McConnell-Ginet, 1999: 186). It is not assumed that such communities are ones where all members share certain assumptions or notions of what is acceptable, for, as Bucholtz remarks, this model ‘treats difference and conflict, not uniformity and consensus, as the ordinary state of affairs’ (Bucholtz, 1999a: 210). Not all members will be equally engaged on particular tasks or enterprises, and not all members will be central to the group but may only play a peripheral role, in terms of their involvement and their affiliation. Furthermore, communities of practice do not exist in isolation, since individuals belong to many different communities and the practices of these groups often affirm or challenge the practices operating in the community of practice, either at a stereotypical or actual level. Thus, what this notion of community of practice enables is a very much more complex notion of the individual who, because of their participation in various communities of practice is constituted as a multifaceted, and potentially internally fractured, self. The individual is not seen as a coherent, stable, entity but as a range of subject positions, some of which come to the fore in interaction with particular groups of other people; the individual engages with others and is defined and changed by that engagement and contributes to the changes taking place within the community of practice.12 Each individual, as well as belonging to many different communities of practice, is also a member of a culture, whether willingly or not. Although the Scollons argue that the term culture itself is

Rethinking linguistic interpretation

31

‘too broad a social organisation to be very useful in the analysis of discourse’, nevertheless, society as a whole has an impact on the linguistic production of individuals just as particular communities of practice do, although perhaps always in a mediated form (Scollon and Scollon, 1995: 168).13 Eelen argues that culture is defined so vaguely in most linguistic accounts that it is unclear what it refers to; furthermore, the conception of culture vacillates between homogeneity and heterogeneity in much linguistic work: ‘If the seemingly ad hoc uses of the term are taken seriously, the notion either annihilates itself, or it annihilates the conceptualisations it is asked to defend. A notion that can simultaneously denote any group of people based on (any combination of) characteristics loses its operational value. On the other hand, if the notion were fully adjusted to the amount of empirical variability encountered, cultures would become so small that the notion of shared norms would lose its explanatory value and fail the explanatory role it is currently asked to fill’ (Eelen, 2001: 173). For him, the problem with the conceptualisation of culture is largely that individuals are seen as simply the passive receptors for cultural values and speech styles, which we paradoxically then trace only in their linguistic behaviour. Whilst it is clear that society is a ‘fictitious body’; in some respects, that does not prevent values which we assume are those of the society as a whole having an impact on us as individuals or as members of groups (Scollon and Scollon, 1995: 207). Foley argues that rather than culture being seen as an abstract set of beliefs, behaviours, traditions, and institutions which transcend the individual, culture should rather be seen as ‘embodied practices’ (Foley, 1997).14 Thus, rather than seeing the relation between the individual and the society, as in some materialist analyses, as one of oppression and indoctrination, Foley is attempting both to see the coercive or affirming force of certain cultural practices and to enable us to perceive a model for change at the level of the individual: in some ways, to reinscribe the agency of the individual without succumbing to the model of the atomistic individual who is in total control of their behaviour. Many of the cross-cultural analyses which have been undertaken into differences in politeness have attempted to make generalisations about cultural predispositions to certain types of behaviour. Politeness seems to be crucial here, as it is around the issue of politeness that a great number of misunderstandings and tensions arise

32

Gender and Politeness

(Hamza, forthcoming). My main aim here is not to analyse exhaustively this cross-cultural research but rather the model of culture which is implicit in much of this work, in order to try to analyse what determines linguistic behaviour, and what role culture, broadly speaking, plays in that process. But I shall not be assuming that culture exists ‘out there’ in any simple form, but that rather I shall characterise it is a set of assumptions made by the individual because of his/her involvement with groups where those values are affirmed and contested. Thus, rather than assuming that certain cultures are negative politeness (concerned with distance) or positive politeness (concerned with involvement) cultures, as Brown and Levinson and Fukushima have, I would prefer to analyse the role such stereotypes of race and culture play in the individual’s linguistic production and interpretation (Brown and Levinson, 1987, 1978; Fukushima, 2000). What must be made clearer in these models of culture drawn on by theorists of politeness is that the ‘culture’ which is described is often the dominant culture, or at least the culture which is assumed to be shared by certain groups within that society, or even perhaps a lowest common denominator of certain types of valued behaviour circulating in the society. Thus, either it is the stereotypical behaviour of groups of males seen to represent the culture as a whole (that is excluding what are seen as minority groups) or it is a synthesis of the stereotypical linguistic behaviour of a range of different groups: thus, the prototypical behaviour of women in certain contexts is melded to the behaviour of men, and that of older speakers is fused with that of younger speakers. To sum up, what I am arguing against is idealised forms of theorising: idealisations of the speaker and hearer; the relation between individuals and groups; and of the model of culture. It might be argued that I am confusing the Model Speaker with actual speakers, and that the linguists who draw on the notion of the Model Speaker are keenly aware of the way that individuals differ in their linguistic production from this idealised model, in much the same as langue differs from parole and competence from performance. However, I would like to interrogate the theoretical purpose of the Model Speaker if it does not approximate to the linguistic production and processing of actual speakers, and furthermore, and perhaps more problematically, I would assert that the linguists who draw on it are in fact often making statements about real speakers and hearers,

Rethinking linguistic interpretation

33

since they often confuse the Model Speaker with real individuals (see Brown and Levinson, 1987). I would like to move away from the use of the notion of the Model Speaker in linguistic analysis to a form of analysis which questions the autonomy of the individual and tries to set him/her in relation to a range of communities of practice where they negotiate their position and their gender, race, and class identities. I would like to question the degree of control which this model of the speaker is assumed to have. Whilst I do not wish to portray the speaker as simply being subject to a range of discursive pressures which determine what is said, it is necessary to be aware of the forces which are at play in shaping the construction and interpretation of utterances by individuals. Model of communication and language Implicit in linguistic accounts is a model of communication. Many linguists assume that it is possible to analyse language in isolation from other behaviour; indeed, as linguists, it is assumed that it is necessary only to analyse language. Some linguists, whilst stating that they will only analyse language, in fact, find it difficult to separate language from other forms of behaviour; for example, Holmes gives as examples of linguistic politeness: ‘avoiding telephoning a colleague on a Sunday morning or apologising for interrupting a speaker . . . sending a birthday card to a friend’ (Holmes, 1995: 5). Whilst these acts are achieved through language, our judgement of them as polite is based on a judgement of them as behaviour rather than as language per se. It is impossible, and politically inexpedient, to see language and behaviour as entirely separate (Beaken, 1996). Toolan argues that ‘whatever language-dedicated cognitive predispositions the normal human may be endowed with – and there is plentiful evidence that humans are predisposed toward the grasp and use of certain kinds of structuring – those predispositions are not, as many have claimed, crucial for language’ (Toolan, 1996: 10). Rather than seeing language as a specific set of abilities or predispositions, Toolan suggests that ‘language draws primarily on quite general characteristics of humans . . . faith, trust, orientedness to others; faculties of memory and imagination; goal orientedness, and the ability to perceive the relatedness and nonrelatedness of phenomena . . . language is essentially a flexible practice, shaped by profound

34

Gender and Politeness

interacting principles of self-awareness, normativity, other orientedness, and rational risk-taking, integral to the larger phenomenon of risk-entailing puzzle-working entailed in life itself’ (Toolan, 1996: 13). Thus, in this form of integrational thinking, Toolan is arguing that language should be set in the context of other forms of human behaviour. There has been a great deal of criticism in recent years of ‘telementation’: the assumption that communication is simply a question of transmitting a message from one individual mind to another. Thus, the speaker has an idea, encodes it in language, and the hearer decodes this message back to the original idea. The process whereby ideas are formulated in and through language is a complex one. Many cultural theorists question the ‘code’ model of language whereby utterances are simply coded by the speaker and decoded by the hearer. Theorists such as Hall suggest that each participant in the process of understanding constructs a slightly different interpretation of the utterance; thus his model is predicated on a difference between the message sent and the message received (Hall, 1980). The Sapir–Whorf model of linguistic determinism sees language as a set of cognitive categories which determine certain types of thought process (Mills, 1997). However, Pinker argues for a notion of mentalese, a form of non-linguistic thought; he suggests that ‘we have all had the experience of uttering or writing a sentence, then stopping and realising that it wasn’t exactly what we meant to say. To have that feeling, there has to be a “what we meant to say”, that is different from what we said’ (Pinker, 1994: 57). ‘What we meant to say’ could still be represented in language, but Pinker argues that ‘sometimes it is not easy to find any words that properly convey a thought. When we hear or read, we usually remember the gist, not the exact words, so there has to be such a thing as a gist that is not the same as a bunch of words. And if thoughts depended on words, how could a new word ever be coined? How could a child learn a word to begin with? How could translation from one language to another be possible?’ (Pinker, 1994: 58). Although this quotation gives rise to many more questions than can be addressed here, it is important in terms of trying to think about the relation between language and thought, and between language and communication. Although strong versions of the Sapir–Whorf hypothesis suggest that language determines that the individual language user will

Rethinking linguistic interpretation

35

express themselves in particular ways or be predisposed to think in certain ways because of the structures available within the language, a weaker, more mediated, model of determinism can be useful in terms of recognising that certain types of cognitive and linguistic structures predispose us to express ourselves in certain ways, through the notion of collective, negotiated linguistic habits or ‘habitus’. Bourdieu’s notion of ‘habitus’ can be seen as the set of dispositions to perform one’s identity in particular ways which are inculcated in the individual by explicit and implicit socialisation (Bucholtz, 1999a, 1999b). Bourdieu describes habitus as ‘the dispositions [which] generate practices, perceptions and attitudes which are “regular” without being consciously co-ordinated or governed by any “rule”’ (Bourdieu, 1991: 12). This set of attitudes or practices which are seen as constituting a norm by individuals is then negotiated with by individuals in terms of their own perception of what is acceptable for their own behaviour. ‘The habitus “orients” their actions and inclinations without strictly determining them. It gives them a “feel for the game”’. And this practical sense of the ‘feel’ of the game, what other people think and what others consider acceptable, ‘should be seen, not as the product of the habitus as such, but as the product of the relation between the habitus on the one hand and the specific social contexts or fields within which individuals act on the other’ (Bourdieu, 1991: 14). Eelen argues that we have to hypothesise a common world, that is a set of beliefs which we imagine exists somewhere in the social world and is accepted by everyone, with which we as individuals need to engage: ‘On the one hand, collective history creates a “common” world in which each individual is embedded. On the other hand, each individual also has a unique individual history and experiences the “common” world from this unique position. The common world is thus never identical for everyone. It is essentially fragmented, distributed over a constellation of unique positions and unique perspectives’ (Eelen, 2001: 223). Thus, one can hypothesise a sense of a shared culture – a set of practices and beliefs which one can describe as one’s culture or society – but this is a hypothesisation. One’s choice of words and the level of formality and politeness which one assumes to be appropriate can be seen as defining one’s position within a group or community of practice: ‘relations of communication – linguistic exchanges – are also relations of

36

Gender and Politeness

symbolic power in which the power relations between speakers or their respective groups are actualised’ (Bourdieu, 1991: 37). A person’s choice of ways of expressing themselves is not a simple reiteration of conventional beliefs, but rather a complex process of hypothesising that ‘common world’ and one’s own stance in relation to that hypothesised position, which is worked out through an assessment of one’s position in particular communities of practice. Bourdieu argues that speakers act as if there were linguistic and behavioural norms circulating within society; however we should rather see this as a process whereby: ‘in reproducing linguistic expressions speakers take into account . . . the market conditions within which their products will be received and valued by others’ (Bourdieu, 1991: 19).15 Habitus should be seen as systems of durable transposable dispositions, structured structures, that is, as principles of the generation and structuring of practices and representations which can be objectively ‘regulated’ and ‘regular’ without in any way being the product of obedience to rules, objectively adapted to their goals without presupposing a conscious aiming at ends or an express mastery of the operations necessary to attain them and, being all this, collectively orchestrated without being the product of the orchestrating action of a conductor. (Bourdieu, 1991: 72)

Although this is a fairly opaque quotation, what Bourdieu seems to be arguing here is that habitus is a flexible system of behaviours which, when engaged with by individuals, perform a structuring role without being ‘invented’ by a single agent or institution. Eelen provides an account of the way that the present range of linguistic behaviour of individuals is based on the ‘creative transformation of present conditions from a position based in past experience’ (Eelen, 2001: 222). Perhaps more importantly is the way that the present behaviour of individuals also constitutes the present conditions for others’ actions, thus forming the conditions for change both at the community-of-practice, societal, and the individual level. For some theorists, and I suspect this may be true for Eelen, the notion of habitus may be a way of moving away from materialist analysis with its stress on the importance of external factors to the individual in selfconstruction, towards a more individualistic framework of analysis. However, I shall be drawing on this notion of habitus together with the model of communities of practice to describe the dynamic way

Rethinking linguistic interpretation

37

in which the relation between individual and the wider social group is figured.16 Implicit in many linguistic accounts of conversation is a model of speech where speakers and hearers agree implicitly to certain types of behaviour and certain rules of what will be assumed to hold during the conversation (Grice, 1975). Grice formulated the notion of the Co-operative Principle and a set of maxims which he asserted were at the basis of conversation and which speaker and hearer both adhere to in understanding each other. Conversation which does not follow these rules, therefore, for these theorists, needs to be understood in relation to motivations for deviation; thus for theorists such as Grice, if we assume that this model of co-operation lies at the heart of conversation, we can account for deviation through inference and implicatures. However, we need not necessarily view this conversational contract as being anything more than an analytical device which helps linguists to describe the way that speakers might make sense of certain but not all utterances. We need not assume that this model of communication underlies the processing by actual speakers and hearers in conversation. Are we to assume that all interactants hold that a norm from which the other speaker will deviate is one of perfect clarity with no superfluity in terms of utterance length? This assumption seems to stem from a particular analytical tradition in philosophy which focuses on the analysis of the proposition – a pared-down utterance whose meaning is explicit. Most utterances are then seen as deviations from this perfect propositional form. Scollon and Scollon argue that Utilitarian discourse privileges the use of anti-rhetorical, empirical language which is free from emotional excess and florid language (Scollon and Scollon, 1995). This pared-down model of language as the minimum propositional value of an utterance assumes that utterances are primarily concerned to impart information in the most effective, efficient way possible. Thus, underlying this model of communication is the notion that the kernel of an utterance is a simple direct statement. Sifianou argues that: ‘the problem is that a great deal of everyday language does not or does not only aim at a maximally effective exchange of information. . . . Thanks and apologies . . . may be perfunctory or sincere but they are usually effective because they fulfil social expectations rather than any conditions relative to truthfulness or

38

Gender and Politeness

brevity’ (Sifianou, 1992: 16). Nor should we assume, as this model does, that the global model of conversation is one where individuals get on well with each other, understand each other completely, and co-operate. Although this may well be the case in certain contexts, understanding and orientedness towards another cannot be taken as more fundamental as a model of conversation than misunderstanding or scheming and self-seeking. What Grice’s and the Speech Act theory’s model of communication omits is our relationship with others: utterances do not only have propositional content: ‘when we communicate with others we simultaneously communicate some amount of information and indicate our current expectations about the relationship between or among participants’ (Scollon and Scollon, 1995: 138). Indeed, I would argue that there are many other things happening in conversation other than relating to others and giving information: achieving long-term and short-term goals, working out those goals, trying to understand ourselves and others, enjoying ourselves, and so on. Most of these elements are not generally considered in the analysis of conversation. Many linguists assume that interaction can be treated as if it were a text, that is, that it is a product rather than a process; however, in conversation, we do not produce ‘text’ in this way; the conversation develops in a much more haphazard way, and we work out what the conversation is about (and sometimes do not work out what it is about) as we go along. Toolan remarks ‘we are constantly making provisional assessments of the current gestalt – of where we are at now and what will likely be understood (what probable sense will be made or taken)’ (Toolan, 1996: 31). Once we move to this more process-oriented view of conversation, where the data we have cannot be viewed as something which has been achieved, but rather something which is still in the process of being worked out, then we shall be less likely to ascribe fixed and stable meanings to individual elements within the data. We can only ascribe parameters to the way that a particular element within the data may function rather than assuming that it is functioning in that way. Drawing on Bateson’s work, Scollon and Scollon argue that ‘every communication must simultaneously communicate two messages: the basic message and the meta-message . . . which indicates how we want someone to take our basic message’ (Scollon and Scollon, 1995: 67). This meta-message

Rethinking linguistic interpretation

39

may not in fact be clearly understood by the hearer. Furthermore, not only is the speaker signalling to the other how to interpret their message but is also showing the other that they understand the conversation so far in particular ways: ‘to maintain coherent discourse, each speaker needs to keep the discourse going while at the same time confirming to the others that he or she has followed what has gone on up to that point’ (Scollon and Scollon, 1995: 68). It is clear from this that interactants do not process language in a word-byword fashion, or even on a turn-by-turn basis, but rather that we operate more with a higher level of processing. When trying to understand someone we continually form hypotheses about the gist, and we try to work out someone’s meaning or intentions through recourse to elements which we judge to be significant in this overall meaning. We try to work out their ‘angle’ or position in relation to what they are saying (i.e. whether they are being polite, friendly, sincere, or impolite, insincere, and threatening) and what our response to that position is (that is, whether we are offended, pleased, and so on). Thus, if someone uses a particular word or construction, we may focus on this in terms of constructing a hypothesised overall slant or gist for their utterances as a whole. Individual words may be taken to act as signs or hints of the overall meaning. This is what Gumperz refers to as ‘contextualisation cues’; he argues that interactants try to ‘channel’ inferences so that certain interpretations are more likely: ‘this channelling of interpretation is effected by conversational implicatures based on conventional co-occurrence expectations between content and surface style. That is, constellations of surface features of message form are the means by which speakers signal and listeners interpret what the activity is; how semantic content is to be understood; and how each sentence relates to what precedes or follows’ (Gumperz, 1982).17 In many ways, we move from the word, the specific, to the global hypothesised meaning, in a to-and-fro movement where each informs our attempts to make sense of utterances.18 If this is the way that we understand and produce utterances, then it seems strange to focus only on the individual words, rather than on the word as an element in a wider making sense of conversation. What many linguistic studies omit from their models of communication is the sense of motives and interests (that is, the sense of ‘what’s in it for me’), and these two elements are essential for

40

Gender and Politeness

understanding why people converse at all, and for understanding why certain linguistic choices are made. In a sense, it is because of the difficulty of accessing the motives and interests of speakers that they are not considered at all in linguistic analyses in general. Toolan stresses: ‘[the] germane factor in relation to which any present use must be seen is what interactants severally take to be the larger purposes of each interactant in the particular exchange: what each participant is after’ (Toolan, 1996: 145). What is also left out of many linguistic models of communication is the sense of the difference between long-term and short-term goals: people in conversation generally have short-term goals however vaguely formulated, that they wish to achieve in the here and now; but they also have longer term goals. For example, a relative of mine often telephones me to ask for small (reciprocal) favours; he always starts the conversation with general questions of conversational interest, his close family, my close family, recent political events, the weather, other relatives, and so on, and only in the last couple of minutes does he broach the subject of the favour. Generally, after twenty minutes of conversation, the small favour is dealt with in two minutes and the call is terminated very quickly thereafter. In this case, the speaker knows exactly what he wants to achieve in the short term, and has this goal structuring the twenty minutes of conversation preceding the request; but he also has in mind the longer term relationship between us, whereby we each have a form of ‘account’ with each other of reciprocal favours which it is in both of our interests to maintain in working order. I am not suggesting that all speakers have such clear goals in conversations (or that my relative has such clear goals in other conversations, or indeed in all conversations with me); indeed, the goals are often only formulated in the process of conversing, but the short-term and long-term interests and motives of conversation are something which is attended to by all participants. Generally, what we are analysing as interactants is not a word-by-word analysis of what someone says, but rather what we think their aim is in saying something: we assume that there are larger purposes at work. Toolan argues that ‘we make sense of a situation, or an utterance, or a “word”, within a speech act, not by breaking complexes down into parts, as if those parts had stable and independent value or meaning, but by postulating (i.e. speculatively composing) the purposive whole of which it is a part’

Rethinking linguistic interpretation

41

(Toolan, 1996: 247). Toolan suggests that the notion of ‘face’, used so extensively in politeness theory to denote the speaker’s self-image which they wish to see maintained within the group, is not adequate to be able to account for this view of speaker motives: ‘face is not nearly strong enough to subsume questions of interest ranging from the petty pecuniary to the politico-ideological’ (Toolan, 1996: 167). We generally have a sense of a range of different purposes which might be at work.19 We also have a variety of time frames in our minds as we speak: we are working out the here and now, which is what Brown and Levinson and most linguists deal with – what is represented – but we may also be thinking about what we want to achieve in the conversation as a whole; what sort of relation we wish to have to particular interactants and the community of practice as a whole. For example, we may be speaking to someone in what we consider to be a positively polite manner, for example, by displaying consideration for their needs, because we know that we shall need their help in the future and we therefore have to be strategic about the amount of attention that we give them. Interactants are constantly trying to assess what is going on at the moment in a particular interaction and constantly revising their assessments of others and what it is possible to achieve.20 In linguistic accounts of language use, particularly those influenced by Speech Act theory, there is an assumption that perfect communication is the norm. Scollon and Scollon, in contrast, suggest that misunderstandings are the only certain thing about conversation, and whilst their discussion is focused on cross-cultural communication, the same should be held true of interaction in general (Scollon and Scollon, 1995). Because interactants are often working out what they want to say in the course of the interaction, and because sometimes our interlocutors may tend to appear obfuscatory to us or at least unclear, not because they are necessarily intending to be unclear but because our frames of reference and theirs are not the same, or because they are having difficulty expressing particular positions or ideas, we make informed guesses as to what they might really mean, and learn to tolerate a fairly high level of misunderstanding or at least a degree of not quite understanding exactly what the other person is saying.21 Wodak, in her book Disorders of Discourse (1998), examines the misunderstandings which take place, largely as a result of structural inequality, between individuals

42

Gender and Politeness

interacting with those in positions of institutionalised power. In her analyses of hospitals and schools, she shows very clearly the effects that differences in power can have on those who are not familiar with institutions. In her analyses of conversation within institutions in place of clarity and comprehension . . . confusion was sown and barriers to communication erected. Such disorders of discourse result from gaps between distinct and insufficiently coincident cognitive worlds: the gulfs that separate insiders from outsiders, members of institutions from clients of those institutions, and elites from the normal citizen uninitiated in the arcana of bureaucratic language and life. (Wodak, 1998: 2)

She gives an example from a dialogue between a doctor and patient in an outpatients’ clinic in Vienna (Wodak, 1998: 1): Patient: Doctor:

What did you mean about cholesterol? Well, there are new guidelines, specifically those regarding HDL and LDL cholesterol. And we’ll take it from you and then we’ll see what happens.

She suggests that what happens in such a dialogue is a ‘frame conflict’ where ‘worlds of knowledge and interests collide with one another and those who possess linguistic as well as institutional power invariably prevail’ (Wodak, 1998: 2).22 Here, the patient seems to be asking for a simple explanation or clarification of the term cholesterol which has been mentioned by the doctor previously, but the doctor responds with information about institutional procedures relating to cholesterol and about specific types of cholesterol testing. Whilst it is necessary to question the model of power used in this form of analysis, since we might question that those in positions of institutional power always determine the overall interpretation of stretches of speech, Wodak’s focus on misunderstanding is instructive, even if we can also see that misunderstanding can arise or be motivated by a wide range of other factors than structural inequality.23 Thus, what I am drawing attention to here is the need to move away from the type of linguistic analysis which characterises interaction as strategic, unmotivated, decontextualised, and readily understandable by all interactants. Instead, I would like to emphasise the way that making sense of conversation is a much more haphazard affair, where interlocutors continually try to make hypotheses about

Rethinking linguistic interpretation

43

what the other person/s means, to gauge their level of commitment to what they say, and to try to construct responses which might be seen to be relevant to previous utterances. Methodology: data collection and interpretation As I mentioned in the Introduction, what constitutes data for analysis is a very complex issue. Quantitative analysis has been subject to a great deal of criticism within linguistics, because of the difficulties of assuming that the language behaviour of people in experimental settings can be generalised to their behaviour in ‘real life’ and to the behaviour of the population as a whole. However, there are other problems associated with quantitative analysis which are largely to do with assumptions which are made about the relationship between language items and their wider functionality. For example, Holmes, in her book on gender and politeness, is forced to assume a functional approach to language in order to undertake large-scale analyses (Holmes, 1995). Thus, she assumes that the analyst can recognise a compliment, characterising compliments as made up of particular types of language items, which can be quantified. As I will be showing in the following chapter, such assumptions have to be questioned, since particular language items are always multifunctional (Cameron et al., 1988).24 Eelen also draws attention to the problems in trying to come to meaningful statistically based generalisations about the language use of a particular community; he suggests that a quantitative analysis will always have to ignore variability within a community in order to assert that a certain type of behaviour is the social norm. He therefore argues that social norms ‘are not derived directly from individual behaviour, but rather from the abstract(ed) social average behaviour. In this sense norms are not empirically found at the individual level, rather they are posited as explanations for the social findings’ (Eelen, 2001: 217) (emphasis in original). Thus quantitative analysis is essentially normative. Those linguists within the field of pragmatics have tended to use qualitative analysis. Thus, samples of conversation are taperecorded from small samples of men and women, white people and black people, and so on, and tentative generalisations about people’s linguistic behaviour is made on the basis of these data. Sometimes, only small groups of interactants are examined, for example, the

44

Gender and Politeness

language of particular groups of women within very tightly defined contexts; no generalisations are thus made about women’s language in general (Coates and Cameron, 1988). However, perhaps qualitative analysis suffers from the same type of problems as quantitative analysis, even though it is more willing to question the possibility of generalising from its findings. One of the difficulties is that often the people drawn on belong to the same linguistic community as the linguist, so there are numerous studies of the language of university students, of middle-class white people, and fewer studies of other groups of people.25 Some linguists, particularly in politeness research, use questionnaires to try and access the type of presuppositions that people make about language. Role-playing exercises are also used to elicit certain types of language. I have no objection to these forms of analysis but have found that, in general, when answering questionnaires, informants tend to provide stereotypical beliefs and language rather than the language that they actually use. Particularly in relation to a subject like politeness where there is often a mismatch between what people think they do, or should do, and what they actually do, the use of questionnaires and role-play should be handled with care. Spencer-Oatey’s edited collection Culturally Speaking is an example of the difficulties in this area; here role-plays, discourse-completion exercises, and gap-fill techniques are used to determine differences in cross-cultural linguistic performance. But, because of the nature of our views about our own position in relation to what we assume our culture is, interactants may feel that they are responding as cultural representatives, and that they are obliged to present positive images of themselves and their culture (Spencer-Oatey, 2000). For example, in Blum-Kulka’s research on Israeli perceptions of politeness, her respondents characterised themselves as polite in stark contrast to those others within the wider cultural group who could be characterised as impolite (Blum-Kulka, 1992). Thus, stereotypes of linguistic behaviour are often drawn on in questionnaires about linguistic performance. Some linguists try to overcome this problem of using real data by using invented examples or introspection. As I discussed in the Introduction, I have decided that anecdotes that people have told me, or my interpretation of events that have happened to me, or that I have witnessed are usable forms of data, alongside other forms of

Rethinking linguistic interpretation

45

data, precisely because of the problems with the notion of objective language analysis. However, these data will be presented as precisely a partial and ‘interested’ form of data, in that in most of the cases, only one interpretation or set of possible interpretations will be available. These anecdotal data are used in this book with theoretical care and for illustrative purposes only. I have tried to incorporate questionnaires and interviews with standard data collection; so that, for example, I tape-recorded conversations and then, after analysing the data, I consulted with the interactants and talked to them about what they thought was going on, playing them back the tape of the interaction, and focusing on particular sections which I felt were significant. This seems to be a better form of analysis as it is concerned with eliciting what the interactants thought was going on. However, it is getting no nearer in essence to what really went on, as it is simply another text, another conversation, only this time the interaction is with the analyst. The complexities of that relationship have to be taken into consideration, particularly when the analyst’s gender, race, and class position are salient, because they differ from (or they are similar to) those of the interactants.26 One of the problems with data collection is what we consider data to be: generally, linguists transcribe tape-recorded conversations, but although they may analyse them for stress, and sometimes intonation, there are many elements which are crucial for understanding conversation which are left out of the analysis. Toolan argues that ‘for integrational linguists, transcription converts an interactional event (or series of events) into a textual product (separable from “the” contextual setting), a property. Transcription is a kind of absconding with that part of an interaction most easily reduced to writing, leaving the remainder as disposable residue’ (Toolan, 1996: 5). Thus, gesture and eye movement are generally not analysed, and the problem with attempting to include these in an analysis, as many Conversation Analysts do, is that we, as analysts, would not know which of these elements was salient for the participants at any particular time: we might record all of the eye movements of participants and yet they may only become important at very specific parts of the conversation, for example, when gazes meet. However, there is no overt marking of this salience by participants (Goodwin, 1981; Atkinson and Heritage, 1984). Similarly, interactants may also be unaware that they have considered eye movement as salient, but that

46

Gender and Politeness

is not to say that it had no effect; for example, in handing on a turn in conversation, or in aligning oneself with another’s argument, eye contact, gesture, and posture may be crucial. Not only are there difficulties in what we assume can count as data but there are problems in how we interpret these data. Analysts often analyse utterances as if the transcription that they have produced is the only way that the conversation could have gone and that their interpretation of the text is both self-evident and inevitable. Whereas, as we all know as language users, there is a fluidity about how conversations go which cannot be grasped in the final transcription of a conversation; there are moments when the conversation seems as if it is pulling in one direction, only to be pulled in another by a particular event or utterance. At these moments, our stores of anecdotes and illustrative stories relating to that theme are either activated or put on hold. But this fluidity of conversation cannot be represented or even accessed by the analyst. Furthermore, the extracts of conversations which are generally analysed in discussions of data are analysed in isolation; the analyst rarely explains what happened after the event, or the way that the event was processed, recycled, or problematised by the participants in later conversations with each other or with others. In interpreting data, one of the debates between CA and CDA is about the role of variables. In CDA, it is generally assumed that context consists of background assumptions, shared conventions, and also variables such as gender, race, class, ethnicity, and so on. Toolan argues that which element makes an impact on the utterance or its interpretation cannot be predicted in advance (Toolan, 1996: 7).27 McClintock has argued that we cannot simply analyse these variables as separate entities which have a ‘bolt-on’ relation to each other, in much the same way as pieces of Lego. Instead, they must be analysed as interacting with each other and inflecting each other in particular ways. McClintock argues that race is always already classed and gendered, just as all the other elements are similarly inflected (McClintock, 1995). Other theorists have been concerned to analyse whether these factors are instantiated and external to the text; ethnomethodologists or those sociolinguists and social psychologists who draw on ethnomethodology argue that trying to analyse conversation with the notion of variables as pre-existent is not adequate for analysis; what is necessary instead is a notion of these

Rethinking linguistic interpretation

47

variables as instantiated in the text. Schegloff (1997) argues that many analysts simply impose their orientation to the subject under discussion onto the participants. However, for him, orientation to a variable like race, class, or gender ‘is not merely the imposition of an external academic or professional analyst, but is the understanding of the co-participant, as revealed in ensuing talk, which is built on just that understanding’ (Schegloff, 1997: 179). Thus, the participants in the conversation make these elements salient. Nevertheless, in the data that I have analysed, class, gender, and racial identity are clearly very salient issues, which interactants orient to within the course of the interaction, but they are also factors which have some determining role to play in the conversation. Interactants are informed by notions of themselves as classed, gendered, and raced individuals as well as being individuals who are related to each other in a community of practice which has particular views of itself and its values. In Butler’s terms, they therefore perform those identities within the interaction (Butler, 1990). Class, for example, is salient in particular ways at particular times; it is clearly constructed in talk, as ethnomethodologists claim, but it is not just simply something which is engaged with in talk as a topic. As Thompson argues, ‘There is an ever present temptation to suppose that class is a thing. By class I understand a historical phenomenon, unifying a number of disparate and seemingly unconnected events, both in the raw material of experience and in consciousness . . . I do not see class as a “structure”, not even as a “category”, but as something which in fact happens (and can be shown to have happened) in human relationships’ (Thompson, 1980: 8). Thus class is a relational phenomenon which ‘happens when some [people] as a result of common experiences . . . feel and articulate the identity of their interests as between themselves and as against other [people] whose interests are different’ (Thompson, 1980: 9). Thus, for Thompson, class is a way of categorising one’s self as having common interests with others, but I would like to add here it is also a way of being categorised by others. Classifying someone as being from the working classes may well be part of a process of excluding that person from access to certain affiliations or privileges. Here, one’s notion of another’s class is based on stereotypes of the working class. Class is also constituted out of differential access to education and assumptions about ‘cultural capital’ (Bourdieu, 1991).28 Thus, variables like

48

Gender and Politeness

class, race, and gender are dependent on an assessment by others and ourselves and those assessments are often dependent on hypothesised stereotypes. If we draw on Bourdieu’s notion of habitus, these variables and stereotypes of race, class, and gender weigh on the production of talk. These stereotypes may also determine the way that others consider us and judge our utterances. However, one’s class, race, and gender position also determine one’s access to certain resources which may affect the type of discourse style which is adopted. Thus, we need a form of analysis which can blend these two approaches together – which both sees variables as something engaged with and oriented to within an interaction (CA) and that which views variables as having a material effect on the production and interpretation of discourse (CDA); it is a form of analysis which is aware of the constraints that class, race, and gender exert on talk, as well as the negotiations which individuals engage with in relation to certain stereotypes of these variables (Liladhar, 2001; Walsh, 2001). Power is another variable which is often treated in a relatively simplistic way in much linguistic analysis. As I shall discuss in more detail in chapter 3 in relation to gender in particular, many theorists only consider institutional status or role as indicative of the individual’s power position. There is also the difficulty that class, race, and gender are often elided with power, so that it is assumed that all women, working-class and Black people in Britain are powerless. Power is viewed as a possession, so that the metaphors which are used are concerned with taking power and depriving someone else of power, as if power were a finite resource. However, as I shall show in chapters 4 and 5, if we use a more Foucauldian and feminist model of power, we shall be able to describe the complexities of power and also the positions from which it is possible and in fact necessary to contest power relations. Since power needs to be renewed regularly through ritual, regular interaction, and particularly through displays of politeness, it is in fact rather an unstable edifice, rather than the stable entity which many theorists assume (Thornborrow, 2002; Manke, 1997). A further problematic issue in the analysis of interaction is what constitutes context. Many linguists reject the notion that context and text are distinct, or that we can distinguish between them in any clear formal way (Toolan, 1996; Duranti and Goodwin, 1992).

Rethinking linguistic interpretation

49

Toolan suggests ‘in our everyday world and our everyday uses of language, we make no clear separation of experience into “texts” and “contexts”’ (Toolan, 1996: 4). He goes on to argue: ‘Context is both indispensable to our making sense of language and shockingly, liberatingly variable; it is only locally determinate, as occasions of communication arise. As a result there really is no such thing as the context, even to the extent that we may continue to think of certain things as texts: there is only a recurrent activity of contextualizing’ (Toolan, 1996: 4). In fact, for Toolan, the division between text and context is simply a post-hoc rationalisation; for participants, what is text and what is context is only decided on provisionally and situationally. As Goodwin and Duranti argue: ‘the dynamic mutability of context is complicated further by the ability of participants to rapidly invoke within the talk of the moment alternative contextual frames’ (Goodwin and Duranti, 1992: 5). And, indeed, different participants may well be operating with different rather than shared contextual frames. Goodwin and Duranti argue that we should see interaction and context as co-constructing as ‘the dynamic socially constitutive properties of context are inescapable since each additional move within the interaction modifies the existing context while creating a new arena for subsequent interaction’ (Goodwin and Duranti, 1992: 5). Schegloff argues that Critical Discourse Analysts are too ready to impose their own notion of what constitutes the context onto interactions: ‘discourse is too often made subservient to contexts not of its participants’ making, but of its analysts’ insistence’ (Schegloff, 1997: 183). However, although context is something which is invoked by speaker and hearer on an ongoing basis throughout the interaction, that does not mean that larger forces are not also at work: ‘this does not mean that context is created from scratch within interaction so that larger cultural and social patterns in a society are ignored . . . Instead . . . even those participants who are strategically rearranging context to further their own goals invoke organisational patterns that have an existence which extends far beyond the local encounter’ (Goodwin and Duranti, 1992: 6). The role of the analyst is often assumed to be neutral in much linguistic theory, especially in those types of linguistics that are trying to emulate scientific models in terms of ratifiability, objectivity, and so on. This analytical position is very much that of an omniscient

50

Gender and Politeness

narrator, unaffected by the vagaries of personal interest or emotion, who can conduct scientific experiments objectively. Within literary and cultural theory, the role of the critic/analyst has come under intense scrutiny. The critic is assumed to be one who has interests and particularities of response to the texts under consideration, and although the critic may well be one who can offer an interpretation of a text which demonstrates a knowledge and awareness of literary modes which is different to many readers, the critic still remains a reader (Mills, 1994a, 1994b).29 This form of questioning of the analyst is important, because it gives the reader/interactant more standing in the analysis; readers are not assumed to be unthinking, non-self-reflexive, beings who are the simple passive recipients of meaning or who accept ideological knowledge unquestioningly. In linguistics particularly, since every speaker and hearer is a skilled user of language to a greater or lesser extent, and since each person is able to comment critically on their own and others’ linguistic performances, we cannot describe linguistic performance from a seemingly neutral or expert position. We need to be aware both of the partiality of our own positions, and also of the meta-language which language users possess. We also need to be aware that the analyst cannot be aware of everything that is going on in the interaction, as Goodwin and Duranti argue: ‘even an observer who has access to a setting and the talk that occurs within it may nonetheless not have access to all of the phenomena that participants are utilising as context for their talk’ (Goodwin and Duranti, 1992: 6). Thus, perhaps linguistic analysis needs to be more modest in what it claims to know about what is going on between interactants. Let us take one example of the tendency in linguistics to assume that the analyst knows what is going on in an interaction. Antaki, in his essay ‘Identity ascriptions in their time and place’ (1998), analyses two extracts of transcribed conversations between a mother and teenage daughter where the daughter uses the phrases ‘Fagin’ and ‘the terminally dim’. Both of these terms clearly have a general meaning which we as analysts and as speakers of English can access – that is, ‘Fagin’ refers to someone who is miserly, and ‘terminally dim’ refers to someone who the interactant wishes to categorise as beyond the pale in some ways, either because of their actual lack of intelligence, or simply because she dislikes them. These are the meanings which Antaki assumes are in play in the conversation; however, in analysing these extracts, Antaki does not allow that

Rethinking linguistic interpretation

51

these two words may have been used before by the participants or that they may be being used in some specific way, so that ‘Fagin’ only really makes sense if the daughter has repeatedly drawn attention to her mother’s seeming miserliness on several occasions before this particular interaction; or if ‘terminally dim’ is a term used by the family to refer to particular types of behaviour, or particular types of people. These terms might also make sense by playing with and reacting to other phrases which are common currency within the family. Thus, rather than this form of ‘close reading’ which assumes that the analyst knows what is going on, we may have to be content to suggest possible scenarios, and we may have to analyse longer stretches of interaction and refer to them in our analysis of shorter extracts.30 Conclusions Whilst I consider that the Model Speaker and the ensuing difficulties which result from its use within linguistics must be fundamentally questioned, I am not arguing in this chapter that all linguistic analysis is therefore suspect and should be ignored. Nor am I proposing a totally new methodology. As Toolan stresses: ‘It is not clear that a radically new methodology is either possible or necessary: it is open to question whether a perspective so definitionally sensitive to the varying requirements of each new communicational situation could yield anything so determinate as a methodology’ (Toolan, 1996: 22). But what I am arguing for is a careful and critical analysis of the type of claims that have been made within linguistics on the basis of this model of speaker and communication. If these claims about what happens in conversation are based on idealised and stereotyped elements, then we should perhaps rethink the nature and generalisability of those claims. By drawing on both CDA and CA, we should instead focus on the forms of sense-making and self-definition which interactants draw on in their negotiations within the constraints of communities of practice and the society as a whole. Notes 1. I should make clear that some linguistic analyses have tried to tackle these issues. What I am trying to map out here is a sense of the general difficulties which are often faced in linguistic research and which I feel need to be resolved or at least addressed in order to develop a model

52

2.

3.

4.

5.

6.

7.

Gender and Politeness of communication and interpretation which is adequate for the analysis of the complexities of gender and politeness. Thus these criticisms will apply to certain fields of linguistic research more than to others. By Conversation Analysts (CA) I mean to refer to theorists such as Schegloff, Atkinson and Heritage, Goodwin, Gumperz, and others, and by Critical Discourse Analysts (CDA), I mean to refer to theorists such as Fairclough, Wodak, and others. The debates between CDA and CA have meant that it is very difficult to combine the openly political stance of Critical Discourse Analysis with some of the analytical sophistication of Conversation Analysis (see Paul McIlvenny, 2002). There are clearly problems with not analysing factors which are seen to be ‘outside’ the interaction, just as there are difficulties with assuming that terms such as gender, race, or class operate in the same way in all interactions, and that power is imposed on interactants. I am not advocating an unproblematised adoption of psychoanalytic theory by linguists. I find theoretical difficulties with psychoanalysis on a number of levels, most notably in relation to the diminishing of the agency of the subject, the role of society in determining certain aspects of the subject’s behaviour, and perhaps more fundamentally, the impossibility, within psychoanalytical models which stress the importance of the unconscious in relation to the subject’s actions, of being able to formulate, bring about, or account for change within the self or the wider society. However, the model of the self developed within psychoanalytical theory is productive in that it enables us to see that the controlling ego is not the only force at work in the production and interpretation of utterances. However, I would argue that ethnomethodologists still assume that there are certain categories which precede analysis; for example, when discussing class and the way that it is played out or worked to within an interaction, the analyst still has a very clear idea of the category of class which they then attempt to show is at work in the interaction. We need to retain a notion of choice and agency on the part of the individual but to see choice as a matter of degree. Many linguists do not describe their model of the self explicitly but in their analyses and interpretations of data it is clear that they assume a self who is strategic and in control (CA); or a self who is at the mercy of social forces and is simply displaying the power relations or gender relations exerted on the interaction (CDA). When analysing the intentions of others, there is a particular problem for linguists when analysing data which have been collected in experimental settings. The Model Speaker has clear intentions which can be easily accessed by the analyst, but this is not the case with experimental subjects. If we are undertaking quantitative research, it is not feasible to analyse carefully the intentions or supposed intentions of each individual, or even the intentions which hearers hypothesise for speakers. In that these hypothesised intentions are in process, fixing on one

Rethinking linguistic interpretation

8.

9.

10.

11.

12.

53

particular moment represents intention as a product. Implicitly when we analyse data there is a sense in which we as linguists assume certain intentions on the part of the speakers, which are predictable and easily derived from linguistic form. However, it is the linguist who imposes that hypothesised intention both on the linguistic form and the speaker. ‘Fighting words’ is a legal phrase used primarily in the US to describe discriminatory language which has to be analysed as intended to offend others because of its sexist or racist nature and which has the force of an action. However, paradoxically, because of Butler’s reliance on Speech Act theory and Althusser’s work on interpellation, whilst trying to approach the question of the institutional nature of these examples of hate speech, she nevertheless draws on a model of the speaker which is very similar to that of the Model Speaker. Particularly with formulaic utterances like certain aspects of social politeness (see ch. 2) such notions of institutional origin and association of utterances must play a role in both the production and interpretation of utterances. In text analysis, Paul Werth has tried to map out the range of assumptions which would have to hold for certain types of text-world to be constructed (Werth, 1999). Although this is challenging work, there are conceptual problems with delimiting which assumptions or knowledge are being brought into play for each element within the text, just as there are in conversations (see Gavins, 2002). In a sense, such a theory has to assume that for the literary text, the same textual world is constructed for every reader and through the same process, and whilst we may be able to describe some of the general processing work undertaken by the reader, the actual text-worlds will differ. They are not so much a property of texts, but a negotiated process between reader and certain hypotheses made on the basis of textual information. There are several difficulties with the notion of a community of practice; one is that it is difficult to delimit and to define. For example, one might ask if the House of Commons is a single community of practice, or is it composed of many different sub-communities of practice which are defined by their position within the House of Commons? Furthermore, we might ask whose definition of a group as a community of practice is accepted; the analyst might describe a particular group as engaged on a task whilst the participants might not consider themselves to be part of a group. A further problem is that focus on the community of practice tends to obviate any consideration of wider social forces; thus gender is considered only at the level of the community of practice and not at a societal or institutional level. The theorisation of communities of practice bears resemblance to the Scollon’s analysis of discourse systems, the discursive structures which define membership of a particular group: ‘each discourse system forms a circle of enclosure which . . . gives identity and security to its members,

54

13.

14. 15.

16.

17.

Gender and Politeness but . . . tends to enclose them within its boundaries, so that it is easier for them to go on talking to each other than it is for them to establish successful communication with those who are outside this circle of the discourse system’ (Scollon and Scollon, 1995: 246). However, they argue that despite this, ‘almost all forms of discourse take place at the intersections of several discourse systems . . . [and] because of the complexity of human social organisation, we believe that communication which takes place exclusively within a single discourse system is the rarest form of communication’ (Scollon and Scollon, 1995: 248–9). Thus, because most communication is across discourse systems, we generally learn to tolerate a certain amount of lack of fit between our own assumptions and those of others. It should be noted however this tolerance is only true of certain communities of practice, as some groups who consider their ways of interpretation to be superior to others may be fairly rigid in the way that they interpret the utterances of others. In many linguistic accounts it is assumed that culture is defined by the boundaries of a language community. However, culture, whilst it consists of a set of linguistic routines defining particular forms of acceptable behaviour, also consists of various institutional and economic constraints on the behaviour of individuals and groups. Linguists, because of their focus on language, sometimes tend to assume that language is all there is, and are unable to see the dialectical relation between language use and the wider societal and economic constraints. Such a notion is related to Bourdieu’s conception of ‘habitus’, which I discuss in more detail later in this chapter. However, feminists working with Bourdieu’s model of language have felt the need to challenge his representation of the speaker as fairly passive in this process, and Bucholtz argues that Bourdieu sees language practices as ‘primarily reproducing existing social arrangements’ (Bucholtz, 1999a: 205). Thus, she proposes a modified version of Bourdieu’s work which stresses the role of individuals as agents in constructing their own sense of identity in relation to particular communities of practice. One of the problems with a more openly materialist Critical Discourse Analysis account is that the description of the role of society in determining the form of interaction is often crude. The role of the state or institutions is often characterised as an imposed repressive power, and the possibilities of resistance and local interactional power and status are not considered. These are issues which I deal with more fully in ch. 4, on Gender. Whilst the notion of contextualisation cues is important, Gumperz seems to stress a model of communication where each participant understands the other perfectly and their contextualisation cues are successful. I prefer a model of communication where these cues are hazarded but not necessarily successful.

Rethinking linguistic interpretation

55

18. What I am trying to indicate is that we operate in a to-and-fro movement, between individual words understood as significant to the whole meaning and a hypothesising of the gist of the overall utterance/turn in order to understand certain words in a particular way. It is thus simultaneously a top-down and bottom-up model of processing. 19. For example, in the above example, both my relative and I enjoy (I assume) engaging in small talk as well as realising that it is a necessary preliminary to other types of talk. Thus, the small talk is an important part of the way our relationship works in general, particularly in relation to others in the wider family, as well as serving a short-term function. 20. The mixed nature of these motives and interests is also generally ignored, since someone may wish to be polite, affectionate, and witty at the same time. Politeness is often analysed as if it were the only motive which leads to an utterance being constructed in a particular form. 21. I recognise that, frequently throughout this chapter, I have used ‘we’ in a very unproblematised way which may seem to some to be analogous to the use of the Model Speaker. However, I should stress that this use of ‘strategic essentialism’, a hypothesised form of behaviour which many people will be able to recognise from the example of anecdotal behaviour, is used to illustrate specific analytical points rather than to prove something in a more general way (Spivak, 1990). 22. Other theorists, such as Thornborrow, question that the interpretation of those in power will always prevail; she analyses resistance in interactions in classrooms and police stations, and even though sometimes, despite resistance, the dominant interpretation of what happened does prevail, she shows clear instances where power is challenged and interpretations are jointly brought about (Thornborrow, 2002). 23. This model of power is also drawn on by other Critical Discourse Analysts such as Fairclough in his work on language and power, in that he assumes that the dominant discourse always prevails (Fairclough, 1989, 1992, 1995, 2000). 24. This multifunctionality of utterances is particularly important in relation to politeness since very often interactants use ambiguous forms so that they can be both on-record and off-record at the same time. 25. Notable exceptions are, of course, linguistic anthropologists, since they largely do not argue for generalisable findings (Foley, 1997). 26. Most linguists assume that the gender, race, or class position of the analyst is only significant if it differs markedly from that of the interviewee; however, any of these elements may also become salient because they are perceived to be shared: thus, certain forms of behaviour may be seen to be permissible, because ‘we are both women’ or ‘we are both white’. 27. The relation between class, race, and gender and their constitution is something which has been much debated within feminist theory and will be discussed more fully in ch. 4.

56

Gender and Politeness

28. One’s relation to, and familiarity with, institutional routines is also important in the constitution of one’s class position. Shaw (2002) and Wodak (1998) have shown how important the ritualised linguistic routines which institutions develop are for denying access to power and resources to certain groups within society. 29. Poststructuralist theorists such as Roland Barthes have stressed the creativity which goes into the act of analysis (Barthes, 1982). 30. However, the notion of how long we have to record interaction for is a difficult one, since interaction is generally open-ended and ongoing. What I am suggesting here is simply that interlocutors have very much in their mind the past history of their interaction with others and this is often brought up in conversations as subject-matter. Analysts therefore have to reflect this quality of interaction rather than treating their transcription of the interaction as a self-contained text which has closure.

2

Theorising politeness

Introduction In this chapter, I critically survey some of the work that has been undertaken on linguistic politeness. Brown and Levinson’s (1978, 1987) model of politeness has influenced almost all of the theoretical and analytical work in this field.1 However, in recent years there have been a number of critiques of their work, many of them synthesised in Eelen’s work (Eelen, 2001). I would like to consider some of the problems which have been identified in their work, in order to suggest alternative forms of analysis. My intention in this chapter and in the book as a whole is not to attempt to negate the importance of this work by Brown and Levinson: in many ways, as a system of analysis, it works very well, within its own terms. However, perhaps this is one of the major difficulties with the model that, in some respects, it works a little too well – it can be made to work well on a range of different languages, as many linguists have shown, but it does this by focusing on a very restricted model of what constitutes politeness (Fukushima, 2000; Sifianou, 1992). And yet, despite the fact that data can be found to fit the model, to prove that speakers use positive politeness and negative politeness strategies, it is clear that politeness is a much more complex phenomenon. Watts et al. argue that ‘politeness, despite the eagerness with which empirical researchers have used existing theories, remains elusive’ (Watts et al., 1992: 11). Although data can be found which seem to prove that this model of politeness is adequate, when we analyse how politeness actually functions within conversation, Brown and Levinson’s model can only deal with certain elements of the data, for example where participants are overtly and clearly polite, 57

58

Gender and Politeness

and not others. My criticism of the model of politeness that Brown and Levinson have developed stems from the difficulties outlined in chapter 1 with the models of communication and interpretation of data drawn on by linguists which lead to generalisations being made about language which are not sustainable.2 In addition, however, I have reservations about the way that Brown and Levinson view politeness itself and the methodology that they employ to analyse it. In this chapter, I review some of the critical literature on politeness, but I do not intend to be comprehensive in my coverage, since many of the theorists are simply applying Brown and Levinson’s model without modification; thus, the criticisms of Brown and Levinson’s work apply equally to their work. However, I shall discuss those critics who have substantially modified elements of Brown and Levinson’s model and have produced new and insightful ways of thinking about politeness. What I should like to propose instead is a far more complex model of politeness which is concerned with the way that assessments of what politeness consists of are developed by individuals engaging with others in communities of practice, in the process of mapping out identities and positions for themselves and others within hierarchies and affiliative networks. I begin by briefly describing Brown and Levinson’s model of politeness. I describe the problems entailed with the use of this model and I go on to examine the problems with what they see as the constituents of politeness, their model of communication, and their methodology. Brown and Levinson argue for a pragmatic analysis of politeness which involves a concentration on the amount of verbal ‘work’ which individual speakers have to perform in their utterances to counteract the force of potential threats to the ‘face’ of the hearer. Face is a term drawn, via Goffman, rather loosely from the Chinese, to describe the self-image which the speaker or hearer would like to see maintained in the interaction (Goffman, 1967, 1999/1967). Brown and Levinson state that ‘face is something that is emotionally invested, and that can be lost, maintained or enhanced, and must be constantly attended to in interaction’ (Brown and Levinson, 1978: 66). A threat to a person’s face is termed a Face Threatening Act (FTA), and they argue that such threats generally require a mitigating statement or some verbal repair (politeness), otherwise a breakdown of communication will ensue.3 They see politeness

Theorising politeness

59

primarily as a matter of strategies adopted by speakers and they analyse four broad strategies: bald on record, where an FTA is presented in unmitigated form; positive politeness, which ‘anoints the face of the addressee by indicating that in some respects, S[peaker] wants H[earer]’s wants (for example, by treating him/her as a member of an in-group, a friend, a person whose wants and personality traits are known and liked)’; negative politeness, which ‘is essentially avoidance-based and consist(s) . . . in assurances that the speaker . . . will not interfere with the addressee’s freedom of action’; and offrecord, where the speaker does not openly state the FTA or does so in an ambiguous way (Brown and Levinson: 1978: 75). Bald on record is characterised as open admission of an FTA; positive politeness is concerned with demonstrating closeness and affiliation (for example, by using compliments); negative politeness is concerned with distance and formality (for example, through the use of apologies, mitigation, and hedges); and off-record is an attempt to avoid overtly committing an FTA, through the use of indirectness, ambiguous utterances, or silence. Thus, politeness is viewed as a form of strategic behaviour which the speaker engages in, weighing up the potential threat to the hearer, the degree of familiarity with the hearer, the power relationship between them, and modifying the utterance accordingly. Problems with Brown and Levinson’s model of politeness This view of politeness is problematic for a number of reasons. Brown and Levinson and many other theorists, particularly Holmes, (1995), whose work I shall discuss more fully in chapter 5, implicitly assume that politeness is necessarily ‘a good thing’, and that politeness is akin to being nice, considerate, and thoughtful. However, this positive view of politeness may be drawn upon by speakers and used in a manipulative, strategic way. Sell argues for a less benign model of politeness, since in earlier periods of British history, politeness ‘would connote, not a refinement in feeling, but only the most sinister refinement in lying’ and politeness should therefore be seen as ‘a velvet glove within which to hide one or another kind of iron fist’ (Sell, 1991, cited in Watts, 1992a: 45). Several of the interviewees for this study remarked upon the way that they felt other people were very manipulative in their use of politeness, that in order to achieve

60

Gender and Politeness

their long-term goals, they were prepared to use politeness ‘insincerely’. In some cases this lack of sincerity is perceived by all of the interactants, for example, the use of the almost mandatory phrase ‘with respect . . .’ or ‘with great respect . . .’ in political debates, used as a preface to devastating criticism. Montgomery (1999) argues that sincerity or lack of sincerity is a key element in judging the validity of others’ statements, and may also be related to the degree to which we feel affiliated to others (see also, Walsh, 2001). In relation to politeness, assessment of the degree of sincerity or commitment of the politeness or impoliteness is crucial. However, within Brown and Levinson’s model we have to assume that all politeness is sincere. Brown and Levinson argue that politeness is essentially a question of avoiding Face Threatening Acts (FTAs). But sometimes their notion of what constitutes an FTA is perverse; for example, they and many other theorists state that asking a stranger to pass you the salt at dinner constitutes an FTA which must be mitigated by the use of ‘Can you . . .’ or ‘please’. Since in most societies, asking someone to pass something constitutes what are considered as ‘free gifts’ and therefore does not threaten face, or risk involving them in further conversation, we might ask what the threat to face in this act consists of: talking to a stranger at all? asking someone to do something which has the potential for refusal? If such acts are considered to be FTAs, then the view of society as a whole is a particularly negative one; and, as Sifianou shows in her analysis of Greek politeness strategies, this seems a particularly Anglocentric view of society and what it is permissible to ask strangers to do (Sifianou, 1992). Furthermore, politeness encompasses a wide range of behaviours apart from simply avoiding a threat to the face wants of others. Politeness may function as a way of avoiding responsibility and it may be used as a way of hiding one’s real intentions. Let us take an example to show the complexity, even within Brown and Levinson’s terms, of deciding what a particular polite act means or functions to achieve. If we analyse apologies, we might be led to assume that an apology, such as ‘I’m sorry for my behaviour last night’, would be an instance of negative politeness which would, in some way, restore the balance for a perceived indiscretion or problem. Brown and Levinson argue that if ‘a breach of face respect occurs, this constitutes a kind of debt that must be made up by positive reparation if the original level of face respect is to be maintained’ and they go on to argue

Theorising politeness

61

that: ‘an apology is a debt that must be paid and cannot simply be annulled by a generous creditor’ (Brown and Levinson, 1978: 241). However, in some senses, an apology itself may also produce a certain degree of imbalance and impose an obligation on the hearer or recipient (Brown and Levinson, 1978: 243). Thus, being polite does not necessarily restore a mythical balance to conversation; it may in fact be judged to be indebting the other party. I would argue that this notion of balance is much more complex than Brown and Levinson propose, and that each interactant might have a different notion of whether the balance has been achieved. Indeed, the notion that balance is achievable may be a working hypothesis in interactions, but does not result in an actual state of balance. Furthermore, whilst the apology may have been sincere, the hearer may decide not to accept it at face value. Lakoff notes that ‘unlike most speech acts it is the form of the apology that counts. It is less important whether it is sincere than that it gets made’ (Lakoff, 2001: 23).4 She goes on to describe the complex positing and avoidance of blame and responsibility which can go on in apologies, for example when US President Reagan was interpreted as suggesting that Martin Luther King had been a Communist, there was a call for an apology to King’s widow. Reagan called Mrs King and she stated that he had offered her an ‘apology’, whereas White House aides stated that Reagan had offered an ‘explanation’. Lakoff asks ‘How was it possible for [Reagan’s] reported conversation with Mrs King to be interpreted by her as an apology and by his spokesman as an explanation? And why was it so important to the Reagan administration that the statement not go into the public record as an apology that they were willing to risk offending a significant segment of the (voting) population?’ (Lakoff, 2001: 26). Similarly, when President Clinton apologised for his sexual misdemeanours, there was great debate about whether his apology was sufficient, and whether the words that he had used could be considered to constitute a full apology. Thus, apologies should be seen as complex negotiations between interactants over status and over who is seen to be ‘in the right’, which cannot be simply analysed as reparations for face threat. The type of model of analysis used by Brown and Levinson reifies politeness. This reification of politeness can be most clearly seen in Leech’s work where he argues that, in addition to Grice’s Co-operative Principle (CP), we should add a Politeness Principle

62

Gender and Politeness

(Leech, 1983). This Politeness Principle consists of maxims of tact, generosity, approbation, modesty, agreement, and sympathy. Leech also suggests that there is a difference between what he terms absolute politeness, that is acts which are inherently polite no matter what the context (for example, offers) and relative politeness, which consists of all of the linguistic acts which are dependent on context for whether they are considered polite or not. Brown and Levinson (1987) disagree with Leech, arguing that the proliferation of maxims in Leech’s work is unhelpful, and that politeness operates in a different way from the CP itself; for them, the CP is an unmarked framework for communication, whereas politeness is a deviation. Some theorists like Jary (1998) argue that, rather than adding another principle to Grice’s CP, we should simply substitute the Principle of Relevance, which he argues, as Sperber and Wilson do, subsumes all of the other maxims (Jary, 1998; Sperber and Wilson, 1986). Jary states: for Brown and Levinson the communication of politeness is the aim of polite linguistic behaviour. But if this is the case, then politeness must always be communicated by the use of what are commonly called polite forms and strategies, or why else would the rational communicator employ them? In contrast, the relevance theoretic view predicts that these forms and strategies will only communicate something above and beyond their underlying message if the value of W [weightiness of the FTA] represented by the form or strategy chosen does not match the communicators’ mutually manifest assumptions concerning W. Brown and Levinson’s account is based on the assumption that there are linguistic forms specified for particular speech acts – imperatives for directives, for example – and that the non-use of these inevitably conveys extra meaning. Sperber and Wilson, in contrast, assume less. Their model rests on the assumption that a communicator will choose the most relevant stimulus compatible with her abilities and preferences and that this will be worth the addressee’s effort to process. (Jary, 1998: 7)

Thus, rather than assuming that politeness is recognised by all interactants, Jary suggests a model of politeness which focuses on the individual processing work required within an interaction and what the addressee might assume of the addresser’s intentions. However, where we need to extend Jary’s and Sperber and Wilson’s work is to see that processing should not simply be seen in terms of the individual’s cognitive processing, as if this takes place in a vacuum. What I am proposing is a model which focuses on the processing

Theorising politeness

63

that an individual does in relation to the norms which s/he assumes exist within the community of practice and wider society. In addition it is important to acknowledge the constraints that those wider groupings impose on the individual. Many critics, drawing on the reified model of politeness adopted by Brown and Levinson, analyse politeness in isolation and do not consider forms of linguistic behaviour related to, and overlapping with, politeness, such as courtesy and etiquette.5 Whilst some imply that etiquette and courtesy are outdated, ritualised, and institutionalised forms of behaviour, not related to everyday politeness, for many of the older middle-class white women whom I interviewed, etiquette and politeness, whilst not synonymous, overlap to a great extent.6 For example, one interviewee stated that she has a rule, which she believes is part of the requirements of behaviour for those of her class background, that she can only leave a party after she has spoken to every person present. These types of ‘rules’ which people think are operating in relation to politeness have some force and therefore should not be separated from politeness as simple ‘folklinguistic notions’ or as ‘outdated practices’ but rather analysed as part of what constitutes social politeness. Watts argues that the history of politeness leads to certain types of politeness and views of politeness itself being ‘fossilised’ or associated with certain positions within society associated with power. He argues that, in the eighteenth century, ‘politeness was inextricably linked to social class and socio-political power, so much so that those who did not cultivate politeness in their own individual styles of language usage were open to social stigmatisation and political persecution’ (Watts, 1992a: 44). He goes on to claim that ‘politeness was a sign of good breeding and high social status, but it did not necessarily correlate with consideration of deference towards other individuals’ (Watts, 1992a: 44). Sell also argues that ‘at the zenith of its lofty meaning, politeness was the quintessentially Augustan aspiration, involving a view of [humans] as both source and beneficiary of the blessings of civilisation and intellectual enlightenment . . . it was associated with the metropolitan aristocracy as opposed to rural life and cultural provinciality. It meant a high degree of mental cultivation and elegant refinement, polished manners and neo-classical good taste’ (Sell, 1992: 110; see also Langford, 1989). This history of a particular type of politeness may or may not have an impact

64

Gender and Politeness

on speakers in the present, but we have to be aware that, for some speakers, the range of class, race, and gender positions available to them determines that the notion of good breeding and social position make certain forms of polite behaviour more salient for them in terms of their self-definition, and stereotypes of gender particularly depend upon these associations (Ehlich, 1992; Sell, 1992). These views of politeness may have very material effects, as Berk-Seligson has pointed out in her analysis of politeness in witness testimony in courtrooms; even when we only analyse the addition of polite markers such as ‘sir’ in the speech of defendants, it seems to make a substantial impact on the outcome of the trial (Berk-Seligson, 1988). It is debatable what the exact function of conventionalised politeness is in courtroom assessments, since it is sometimes argued that politeness (or at least such factors as deference and hesitation) are indicative of powerless speech (O’Barr and Atkins, 1980). However, Parkinson (1979) argues that ‘defendants who use polite forms and speak in complete sentences are more likely to be acquitted’ (cited in Berk-Seligson, 1988: 413). Since middle-class speech norms tend to predominate within the courtroom, it is highly likely that questions of etiquette and formal politeness play a role in constructing a positive and hence sympathetic view of defendants. Thus, associated elements such as etiquette and courtesy may play a role in individual assessments of what level or type of politeness is appropriate in a particular context, and therefore should not be excluded from our model of politeness. Brown and Levinson vary between seeing human beings as, in essence, co-operative towards one another – politeness thus functioning as a signal to both participants of this co-operative intent – whilst at the same time holding to a view that humans mask their aggression towards one another through the use of politeness, where politeness functions as a protective barrier between interactants.7 In the first view, politeness is a signal of co-operativeness; however, Jucker argues that such a view of politeness is not general enough, because it ‘applies to certain types of verbal interaction only . . . [it] applies only to those that are co-operative. For instance, communication in police interrogations, in cross-examinations in court, in political interviews, etc, are, as is well known, very often less than optimally co-operative’ (Jucker, 1988: 376). Werkhofer criticises the second view that politeness functions to protect individuals from

Theorising politeness

65

others, when he says: ‘polite conversation is defined [in Brown and Levinson’s work] as a way of avoiding or compensating for the undesirable consequences of “true” communication, or even of breaking communication entirely’ (Werkhofer, 1992: 181). Both views characterise politeness as unilaterally functioning to achieve one particular goal in interaction. Sifianou tries to see politeness less as a means of ‘restraining feelings and emotions in order to avoid conflict’ and more as ‘a means of expressing them’ (Sifianou, 1992: 82). Thus, in her view, politeness allows aggression a socially acceptable or mediated form of expression. Brown and Levinson, in a similar fashion, argue that ‘politeness, like formal diplomatic protocol . . . presupposes that potential for aggression as it seeks to disarm it, and makes possible communication between potentially aggressive parties’ (Brown and Levinson, 1987: 2). However, this view of communication, that human beings are, in essence, aggressive towards each other, and that politeness is what mediates this ‘primitive’ aggressive drive, is ideological. We need not see participants in interaction as either primarily aggressive but mitigating their behaviour for their own ends, or as primarily co-operative and only violating that co-operative principle for specific aggressive ends. Interactants in conversation at different times in a conversation, according to their own perceived needs and pressures upon them will be aggressive or co-operative (or many other types of behaviour), but neither one needs to be seen as primary. Werkhofer (1992) suggests that, instead of attributing a particular value or function to politeness, we see it as a medium like money which mediates between individuals but which does not have any particular force or value itself; like money, it is only important for what can be achieved through its use.8 Thus, what I should like to integrate into Brown and Levinson’s model of politeness is a sense of variability among speakers, and speakers negotiating with a range of constraints that are determined by the context and social pressures of the community of practice as a whole. For example, for some of the older female, white, middleclass women whom I interviewed, politeness was seen by them to be their ‘job’, when they were in a group conversation, particularly for those who did not have paid employment outside the home. For example, if one member of a group was interrupted by someone and thus prevented from completing a story or a joke, several of these interviewees remarked that they felt that they should ‘sort it out’

66

Gender and Politeness

and ensure that they asked the person at the next possible opportunity to continue their story. Indeed, having observed all of these female interviewees in conversations in which I have taken part, this is precisely what they consistently do. Part of their perceived role is keeping a watchful eye on the way conversation proceeds, ensuring that everyone has a fair share of speaking time and is not interrupted or silenced, particularly by male speakers. Watts et al. remark: ‘Politeness is a dynamic concept, always open to adaptation and change in any group, in any age and indeed at any time. It is not a social anthropological given which can simply be applied to the analysis of social interaction, but actually arises out of that interaction’ (Watts et al., 1992: 11). However, whilst this is true, what must be recognised is that within British culture at present, many white, middle-class females regard politeness and etiquette as their occupation and many of them feel it is of great importance; however, this view of the importance of their linguistic work is not shared by everyone and many in other social groups see politeness as superfluous and trivial, perhaps because of its association with this group of women. Brown and Levinson concentrate on the analysis of the individual speaker’s intentions in relation to politeness, and this focus on the individual seems to contradict their characterisation of the phenomenon of politeness itself which is, seemingly for them, so groupand Other-oriented. As Ehlich argues ‘polite activity is an activity that recognises the socially constructed limit as being relevant to the activity itself’ (Ehlich, 1992: 76). Held argues that Brown and Levinson manage to deal with the relationship between the group and the individual in that their model ‘[reveals] and [abstracts] out the multi-level relationship of tension between universality and specificity on the one hand and strict conventions and situationspecific variation on the other’ (Held, 1992: 131). However, she does recognise that, in their model in general, ‘the broad scope of polite behaviour has . . . undergone a certain reduction to rational, goal-directed behaviour strategies’ (Held, 1992: 131). This focus by Brown and Levinson on the individual strategies does not allow us to analyse the way in which individuals are constrained in their behaviour because of expectations which they assume operate in the community of practice.9 If we take the example of formulaic politeness elements, such as ‘please’ and ‘thank you’ and greeting

Theorising politeness

67

someone when you meet them, these elements are not decided upon by the individual at all (although there may be some leeway in terms of personal style). As Harris Bond et al. argue, ‘When particular linguistic items are frequently used to perform a particular communicative strategy, they become conventionally associated with that strategy . . . the speaker who uses one of these will not be taken to have communicated anything about his or her politeness, but rather simply to have fulfilled a social convention’ (Harris Bond et al., 2000: 68). Thus, certain language items which may be classified as polite within the Brown and Levinson model are generally not perceived by interlocutors as polite at all, but simply as behaviour which is demanded by the context. For example, there is a range of behaviour which is generally performed ‘just to be polite’ or ‘for politeness’s sake’. When I offered a good friend some marrows, as we had a glut, she said ‘I’ll refuse and be rude, rather than accept and throw them away, as I don’t like marrows.’ This statement suggests that certain people would accept the marrows ‘out of politeness’, even though they did not want them, and would then throw them away, because refusing a gift could be interpreted as impolite. Here, the fact that the person is a close friend and also that she has a consistently direct personal style enables her to ‘override’ such considerations of the social obligations of politeness. But others may be constrained to perform acts which are not necessarily in their interests but which they do because they feel that the situation demands it. This sense of behaving ‘for politeness’s sake’ cannot be captured within Brown and Levinson’s model. A further example of this social politeness can be found in Christie’s work, where she discusses the use of the third person pronoun and honorific terms in Parliamentary discourse, where reference to ‘the honourable member for Slough’ is determined by conformity to the norms associated with the context of the House of Commons, rather than being an individual choice on the part of the speaker (Christie, 2002; see also Shaw, 2002). Janney and Arndt suggest that we therefore need to integrate into Brown and Levinson’s model a notion of ‘social politeness’ which is ‘rooted in people’s need for smoothly organised interaction with other members of their group. As members of groups, people must behave in more or less predictable ways in order to achieve social coordination and sustain communication’ (Janney and Arndt, 1992: 23). Whilst I would not agree with their emphasis on the smoothness

68

Gender and Politeness

of interaction as the aim of interactants, I believe this notion of social politeness is useful, enabling us to see communities of practice providing a framework within which individuals come to judge what they see as ‘appropriate’ linguistic behaviour. What Janney and Arndt distinguish between is ‘social politeness’ and ‘tact’, that is, a sense of a level of politeness which the speakers and hearers assume is determined by the context, setting, or wider social rules (social politeness), in contrast to the individual choice about what level of politeness to use (tact).10 Janney and Arndt state that ‘social politeness is somewhat like a system of social traffic rules, while tact is more a matter of interpersonal driving styles and strategies’ (Janney and Arndt, 1992: 24).11 Ide makes a similar distinction between ‘discernment’ and ‘volition’, that is those elements which are determined largely by the language and culture, and those which are chosen by the speaker her/himself (Ide et al., 1992). In cultures such as Japan, where honorifics are part of the grammatical structure of the language, and where their use is seen as showing respect and recognition of one’s position within the social order, Coulmas argues that ‘[honorifics] are an essential part of linguistic conduct at all times, in most cases because their omission would render the utterance incomprehensible or provoke misunderstanding’ (Coulmas, 1992: 320).12 In cases such as this, the expression of linguistic politeness and the force of social imperatives seem to be remarkably interlinked. As Fukushima remarks: ‘Brown and Levinson see politeness as an instrumental system of means to satisfy individual face wants, while the Chinese view politeness as exercising a normative function in constraining individual speech acts as well as the sequence of talk exchanges. Failure to observe politeness will incur social sanctions’ (Fukushima, 2000: 52). This is also true not only of the system in China and Japan but also in Britain, where failure to adhere to what is socially expected also leads to social sanctions. Watts argues that we need to make a distinction between ‘politic’ and ‘polite’ behaviour; for him, politic behaviour is ‘socio-culturally determined behaviour directed towards the goal of establishing and/or maintaining in a state of equilibrium the personal relationships between the individuals of a social group’, whereas polite behaviour is ‘making other people have a better opinion of oneself’ (Watts, 1992a: 50). This seems a useful, though counter-intuitive, distinction since in some research on politeness it is in fact social

Theorising politeness

69

politeness or politic behaviour which is taken to stand for politeness in general. For some of the children I interviewed, politeness was considered to be saying ‘please’ and ‘thank you’ and not swearing, and they mentioned that it was concerned with ‘being nice’ or ‘considerate’. Thus, politeness for them consisted of those behaviours which had been drilled into them by adults and which social convention at home and at school dictated that they should use. Interestingly, it is in politic/social politeness, that form of conventionalised politeness which needs to be taught explicitly to children, that there was a great deal of resistance and a great deal of variation. Different children within the group I interviewed had varying ideas of what exactly was acceptable and appropriate even within social politeness, which Watts (1992a) suggests implicitly is largely a matter of socially agreed norms.13 The notion that one could accrue some positive benefit for oneself through the use of politeness was commented upon only by older interviewees and seems to be a more sophisticated and manipulative view of the possible function of politeness. Watts (1992a) argues that verbal acts are assessed as polite only when they go beyond the bounds of what is considered appropriately politic; for example, when someone uses ‘sir’ in addressing an interlocutor, when strictly speaking, it would not be considered necessary within that particular context and community of practice.14 Thus, through this type of behaviour, the speaker manipulates the perceptions of the norms of polite behaviour in order to enhance his/her standing in the group. Watts suggests that only behaviour which can be classified as ‘non-altruistic and clearly egocentric’ should be termed polite behaviour, in sharp contrast to conventional ways of viewing polite behaviour as displaying concern for others (Watts, 1992a: 69).15 For Blum Kulka, what is important about politeness is that it is a form of behaviour which others judge us on globally; she asks ‘Why be linguistically polite? In other words, why do languages around the world provide their speakers with alternative modes of expression for both propositional and relational attitudes, assigning social values to their choices?’ (Blum Kulka, 1992: 270). Although this is not a question which can be answered, since the origins or development of politeness are not accessible to inquiry, the question is a valid one which forces us to consider the exact range of functions of politeness.16 Politeness cannot simply be seen as a form of

70

Gender and Politeness

behaviour chosen by individuals for reasons of considerateness for others, for reasons of self-interest, or because of social constraints, but must be seen as a type of behaviour which may be chosen or which we may feel is forced upon us, for a range of different motivations. This multifunctionality helps to explain the wide range of interpretations which may be given to utterances intended as polite by others.17 In considering social politeness, it is important to consider the notion of appropriateness. Appropriateness is very difficult to engage with, as Walsh has noted, since it is often associated with linguistically conservative analysis, denoting what should be considered appropriate (Walsh, 2001: 9ff.). Appropriateness, or appropriacy, is a term which is generally employed to avoid analysis of the structural inequalities in conversation which lead to certain modes of evaluation being drawn on which favour the dominant group’s norms. Janney and Arndt argue that ‘as long as politeness is defined as (linguistically or conventionally) “appropriate behaviour” . . . little in the way of an adequate approach to the subject is likely to emerge. Our suggestion [is] . . . to lower the level of idealisation, leave the analysis of the rules of politeness (and other logical constructs) to philosophers, and begin paying more systematic attention to how people actually express their feelings to each other in everyday conversation’ (Janney and Arndt, 1992: 22). The problem arises if we assume that it is possible independently to assess appropriacy; however appropriacy is something which individuals formulate themselves in order to judge others’ and their own utterances. Appropriateness remains a useful term to use with caution when discussing the way that individuals come to an assessment of their own and others’ utterances in relation to a set of perceived group norms. It should be noted that individuals may have misguided notions of what is appropriate within a particular group, for example if they are a peripheral group member (Bucholtz, 1999b). Individuals may also not themselves decide upon what they consider appropriate but simply follow the lead of others. Thus, if one member of the group adopts a particular style and level of politeness, for example, using title and last name, other people may follow this practice (Bargiela et al., 2002). Furthermore, some of the people I interviewed remarked on the way that they felt that the politeness norms of a particular community of practice conflicted with their

Theorising politeness

71

individual wishes or desires. These hypothesised norms determined what for them was appropriate behaviour. For example, several interviewees remarked on the way that, in certain conversations, they continued to talk on particular topics which they themselves were not interested in, simply because the person they were talking to seemed to be interested, and there was no easy way to change the topic to something more interesting. One interviewee remarked that at a party which had been given in her honour, she found herself talking to someone whom she was not particularly interested in, who had been rather left out of the general conversation. However, because she had been talking to him for a certain length of time, she felt she could not find a polite way to leave him and circulate, as she had wished. The politeness norms of the community of practice of the party, which determined that she should circulate, clashed with her perception of immediate politeness norms which made it difficult for her to leave this person without an adequate explanation. Other interviewees remarked on the way that fear of silence within the British context was responsible for them speaking more than they might wish to in certain circumstances, when others were not contributing to the conversation. They regarded continuing to speak as polite and appropriate linguistic behaviour. I shall thus be retaining the notion of appropriateness, but it will be a form of hypothesised appropriateness which is not simply externally defined. Drawing on Bourdieu’s (1991) notion of habitus, I argue that this sense of appropriateness is one which varies slightly from speaker to speaker; so that rather than appropriateness being imposed by society, or by the speech community of practice or class, or even by the context, appropriateness is something which each individual has to work out, by assessing their own status in relation to other participants in the community of practice, and by assessing what they think the context demands. This means that they constantly have to assess their own position and identity/role within the group in order to evaluate what is appropriate for them and others, and to assess whether they are going to abide by these rules or flout them.18 This notion of hypothesised appropriateness is important as it entails seeing politeness not as intrinsic to particular speech acts but rather as a process of assessment and judgement. Eelen argues that within Brown and Levinson’s model: ‘politeness is regarded as a unique and objective system that exists “out there” in reality, that

72

Gender and Politeness

can be discovered, manipulated and examined just as any physical object can’ (Eelen, 2001: 179). He argues that interactants behave as if these norms exist and as if politeness were a product rather than a process; however, rather than assuming that these norms do in fact exist, we as analysts need to examine the way that, throughout conversations, participants assess whether the utterances of the other interactants can be classified as polite or impolite, according to a range of different hypothesised norms. We must also be aware that judgement of politeness is something about which there is conflict. As Ehlich remarks: ‘[politeness] is derived post actionem as the result of a process of judgement, and this in turn takes account of a standard lying beyond the action itself . . . In order to be able to qualify politeness as such we need to know what constitutes the standard . . . At the same time we need to know what constitutes the evaluative competence’ (Ehlich, 1992: 76). This judgement of politeness has to do with whether it can be assumed that the other participants are acting in good faith, which is not exactly synonymous with the notion of ‘face’ or ‘co-operativeness’. In a sense, interactants marshal evidence as to whether other participants are operating with their own and others’ interests in mind, and a variety of ‘signs’ may be called upon in order to claim that someone is, or is not, being polite globally. Brown and Levinson themselves (1987) draw attention to the way that the judgement of politeness is crucial; they show that people participating in experiments have been shown to rate politeness in different ways from that anticipated by their model; for example, in some experiments, there were challenges to the notion that offrecord is more polite than indirect, with most subjects identifying indirect as the most polite. Brown and Levinson note: ‘when utterances constructed like hints are actually, in the context, on record (that is, when only one interpretation is acceptable in the context) they are sometimes positively polite (as with irony or understatement) and sometimes negatively polite (as with indirect speech acts)’ (Brown and Levinson, 1987: 20). However, it is difficult to see how it is that these strategies can be clearly categorised at all, since it is a question of the judgement on the part of the hearer alone which will finally disambiguate them, or perhaps discussion between the speaker and hearer. Thus, what we need to emphasise is that, in contrast to Brown and Levinson’s model of politeness as strategic

Theorising politeness

73

behaviour on the part of the speaker, politeness is in fact a question of judgement of utterances in relation to a hypothesised appropriateness. A further example of the ways in which politeness should be seen as not simply displaying concern or respect for others, is when politeness strategies are used in a slightly coercive way, for example, Paoletti (1998) remarks on the way that those in hospitals working with older Alzheimer’s patients tend to complete the sentences of those whom they assume are incoherent, and do not seek clarification and request further information, since this would draw attention to the problems of coherence, memory loss, and intelligibility associated with Alzheimer’s. With younger patients ‘any incoherence . . . displayed [is] treated as accountable, prompting requests for clarification and explanation’ (Paoletti, 1998: 189). Grainger has also noted the problematic function of humour used by nurses to the elderly patients in a geriatric ward, since this seemingly positive politeness may be interpreted as infantilising and offensive by the patients (Grainger, 2002). Thus, although the strategy of completing someone’s utterance or being humorous may be motivated by questions of consideration, in fact this type of politeness can be viewed by patients or relatives as patronising and ultimately as constructing a role for the elderly which they themselves might reject. Politeness can also be judged, by certain groups, to be trivial and a waste of time, as Hamann notes ‘a little ill-manneredness is still more acceptable than empty, polite prattle’ (cited in Ehlich, 1992: 101). Holmes argues that ‘males and females have different perceptions of politeness, where women consider politeness to be of great importance, whilst in general, men appear to feel politeness is dispensable between intimates in private. In some public spheres . . . men seem to regard politeness as unnecessary’ (Holmes, 1995: 194). As I show in chapter 5 on gender and politeness, it is questionable whether all men consider politeness to be inconsequential and all women view politeness as of value; however, this assertion is interesting in identifying stereotypical beliefs about the association of politeness with the private sphere (and hence with women) and direct, informational talk with the public sphere (and hence with men). The feeling that politeness is empty talk and insincere, whereas direct statement of intentions and feelings is preferable, is something which many of the interviewees remarked upon, although it is not

74

Gender and Politeness

always something which is carried over into their linguistic performance. Blum Kulka’s Israeli interviewees viewed politeness, far more than Anglo-American interviewees, as irrelevant and perhaps even manipulative, whereas direct speech was valued, partly, she argues, because Israeli Jews reject modes of speech associated with Europe and the Old World (Blum Kulka, 1992).19 Thus, politeness can be seen to have a wide range of meanings, both for theorists and interactants, and can be used to describe a very diverse set of behaviours: ‘politeness spans the full range from deliberate, conscious linguistic choices to the unconscious application of rules or scripts, as well as the unmarked (politeness as the normal, usual unnoticed way of interacting) to the explicitly marked (for example, Watts’ notion of politeness as “more than merely politic”)’ (Eelen, 2001: 23). Interactants themselves take a variety of positions on politeness itself, some viewing it in positive terms and some treating it with contempt, and this again may depend on the particular community of practice with which they are interacting at that moment and with whom they are aligning themselves. I should like to move politeness research away from the analysis of necessarily strategic behaviour on the part of individual speakers, current in much theoretical work, to an analysis which views politeness as a practice enacted within a community of practice with all the gender, race, and class constraints on linguistic behaviour that this entails, and which also stresses the flexibility and variability of the assessment of politeness from group to group and from person to person. Problems with the constituents of politeness As well as problems with Brown and Levinson’s model and definition of politeness, there are also theoretical problems with the elements which they describe as constituting politeness. Here I consider Brown and Levinson’s analysis of strategic politeness, positive and negative politeness, face and Face Threatening Acts (FTAs). As I mentioned in the previous section, for Brown and Levinson and many other theorists, politeness is a form of behaviour which individuals decide upon, which is used strategically by them. They discuss politeness in terms of strategies and super-strategies, where people think first and then act (Eelen, 2001). They list the acts which

Theorising politeness

75

can be considered positive or negative politeness, which attend to the positive and negative face wants of the interactants. Thus, positive politeness strategies stress the extent to which the speaker and hearer share similar interests and are part of an ‘in-group’, whereas negative politeness strategies aim to demonstrate that the speaker recognises social distance and does not wish to impose on the hearer. Scollon and Scollon analyse these positive and negative strategies and argue that they should, instead, be termed ‘involvement’ and ‘distancing’ strategies, since this avoids the evaluation implied in Brown and Levinson’s terms (Scollon and Scollon, 1995). However, as several critics have shown, involvement strategies in the Scollons’ terms and positive politeness in Brown and Levinson’s terms are not always interpreted as being polite by hearers, particularly in cross-cultural interactions (Bargiela et al., 2002; Spencer-Oatey, 2000). Brown and Levinson argue that positive, negative, and off-record super-strategies can be seen to be in ranked order, with off-record being the most face-redressive, followed by negative, and then positive politeness. This view has been criticised by critics such as BlumKulka, who, when analysing data gathered from questionnaires to Israeli respondents, found that there was no clear ranking of these strategies (Blum-Kulka, 1992). Sifianou argues that where indirect and off-record utterances are conventionalised within a culture, they should not be regarded as more polite than other forms of politeness: ‘members who use indirect utterances . . . must share certain knowledge with the other members of their group which guarantees correct interpretation and success. If that is the case, the process of interpretation is not lengthier and there are actually no more options really open to the addressee/s but to conform to the request, than there would have been had the speaker used a different construction’ (Sifianou, 1992: 119). Thus, cultural norms make indirectness the norm within British culture and therefore it is not any more polite in itself than the use of directness in other cultures, where indirectness is not the norm. For example, in Moroccan Arabic, if you wished a member of your family to bring you an ashtray, you would say ‘Jeeb liya tafaiya’ (Bring me an ashtray); any indirectness, for example using a phrase concerning the ability of the interlocutor to perform the act as in the English ‘Can you/Could you’, would be considered impolite, because you would be deemed to have assessed your relationship with the interlocutor incorrectly.

76

Gender and Politeness

If we analyse politeness as strategic, we might well ask along with Brown and Levinson: ‘why then, given the dangers associated with FTAs, do actors not take out the maximum insurance policy and always choose the off-record strategy?’ (Brown and Levinson, 1978: 79). They argue that because of the ambiguity of off-record strategies, the risk to comprehension outweighs the need to pay attention to face wants. They draw attention to the potential dangers in not assessing FTAs appropriately: ‘if an actor uses a strategy appropriate to a high risk for an FTA of less risk, others will assume the FTA was greater than it in fact was’ (Brown and Levinson, 1978: 79). However, rather than seeing politeness as simply strategic, we need to consider politeness as the result of a range of different factors: it is clearly chosen strategically in some contexts by speakers, it is also used as part of the ‘face work’ that speakers perform, but it is also determined by their assessment of setting or the context and by personal habit or style, as I discuss later in this chapter. Negative and positive politeness are generally characterised in Brown and Levinson’s work as diametrically opposed strategies, but in several points in their work they seem close to acknowledging that they are not so much opposite tendencies but different in kind. For example, they comment on the Scollons’ work on the Athabaskan Indians in Canada: positive politeness, which is relevant to all aspects of a person’s positive face, is a quite different phenomenon from negative politeness, which is specific for the particular FTA in hand. [The Scollons] argue that positive politeness is naturally escalated in an interaction (a positively polite utterance is naturally responded to by one upgrading the degree of positive politeness) and hence unstable; in contrast negative politeness, lacking the escalating feedback loop, tends to be stable, suggesting (implicitly) that these two super-strategies cannot be ranked on a unidimensional scale. (Brown and Levinson, 1987: 18)

Thus, Brown and Levinson seem to recognise that positive and negative strategies are different sorts of behaviour which, perhaps, need different forms of analysis. Other critics, such as Werkhofer, comment on this difference in function: ‘whilst positive politeness . . . is relevant to all aspects of a person’s positive face, negative politeness . . . is specific to the FTA in hand. That is to say that it is only negative politeness that is strictly seen to be redressive of an act which threatens [face] . . . Positive and negative politeness emanate from

Theorising politeness

77

different intentions and lack functional “sameness”’ (Werkhofer, 1992: 179). Harris (2001a, 2001b) also questions the notion that negative and positive politeness strategies should be seen as polar opposites. In her work on Parliamentary debate, she finds that elements of positive and negative politeness are employed at the same time, within the same utterance. The way that the notion of ‘face’ is used in discussions of politeness is also problematic. Bargiela (2000) argues that Brown and Levinson have misread Goffman, who in turn has over-emphasised certain elements from the original Chinese conception of face. The Chinese notion of face is not the same as the individualistic notion of face that Brown and Levinson use. Rather, it consists of two elements: mianzi – reputation/prestige; and lian – respect of the group for the person with a good moral reputation. Mianzi is something which is striven for, whereas lian is ascribed face (Fukushima, 2000). Thus, face, as a whole, consists of a concern that one is conforming to or aspiring to a position in the group or wider society. Brown and Levinson stress the necessity to include the analysis of the relationship between the individual and the group, since they state that there is ‘mutual vulnerability of face’ and that everyone’s face depends on everyone maintaining everyone else’s face’ (Brown and Levinson, 1978: 66). However, when the computation of negative and positive face is analysed, it is only the individual’s face wants which are considered, and they are considered in a fairly negative way. Thus, negative face is considered to be freedom from imposition by others and positive face is the self-image which is approved of by others. Several theorists have criticised both the over-extension and the limitation of use of the term ‘face’ in Brown and Levinson’s use (Bargiela, 2002).20 The notion of face is not adequate to encompass the negotiations between people in conversations: although it covers the details of managing harmonious relationships, it does not deal with the negotiation of interests, manipulativeness, external pressures, and one’s relation to the community of practice which I discussed in chapter 1, which are of great importance in terms of people’s decisions about what type of language they consider to be appropriate in a conversation. Held, for example, argues that fear is perhaps a more adequate notion than face, for she states that ‘the fear of disharmony in relationships, or the charge of wrong behaviour, of unjustified claims for self-realisation, a fear that the

78

Gender and Politeness

other person might “bite back” is more of a defining issue than that of face alone’ (Held, 1992: 145). Spencer-Oatey suggests that we should not use the term face, and use the term ‘rapport-management’ instead; she states ‘the term “face” seems to focus on concerns for self, whereas rapportmanagement suggests more of a balance between self and other. The concern of rapport-management is also broader; it examines the way that language is used to construct, maintain and/or threaten social relationships and . . . includes the management of sociality rights as well as of face’ (Spencer-Oatey, 2000: 12). For many cultural groups, loss of face refers primarily to problems over the perceptions of an individual group member’s place in the social system, which is of importance for the functioning of the group as a whole, rather than in terms of individual face loss. Brown and Levinson distinguish between positive face and negative face, the concern that one is appreciated by others and the concern that one is not imposed on by others. Spencer-Oatey questions this view, since she maintains that ‘Brown and Levinson’s . . . conceptualisation of positive face is underspecified . . . and the concerns they identify as negative face issues are not necessarily face issues at all. I propose instead that rapport-management (the management of harmony–disharmony among people) involves two main components: the management of face and the management of sociality rights’ (Spencer-Oatey, 2000: 13). She argues that negative face is, in fact, a concern for sociality rights. Thus, face rights are concerned with the individual and the respect which is accorded her/him by others, and sociality rights ‘are concerned with personal/social expectancies and reflect people’s concerns over fairness, consideration, social inclusion/exclusion, and so on’ (Spencer-Oatey, 2000: 14). She argues that face itself needs to be subdivided into quality face, that is, the value which is attributed to you within the group; and identity face, which is the role which you have been accorded. Spencer-Oatey also considers it necessary to reconsider the conceptualisation of the FTA, as she argues that FTAs are not simply about a threat to someone’s self-image but also can be considered as ‘rights-threatening behaviour’, that is, they are about the general conception of what is appropriate and fair behaviour within a group (Spencer-Oatey, 2000: 16). Again, her analysis focuses more on the social aspect of politeness than Brown and Levinson’s model does.

Theorising politeness

79

This modification by Spencer-Oatey of Brown and Levinson’s work on face is important in making more explicit the role of the community of practice and wider social grouping in perceptions of face. As I mentioned earlier in this chapter, I do not consider politeness to be simply about the avoidance of FTAs. Since Brown and Levinson’s model is centred on FTAs, the instances where politeness is not FTA avoidance or mitigation are not considered in their work. It is important to note that politeness, even when it is associated with FTAs, still allows the FTA to be performed; it does not erase the effect of the FTA. For example, when a builder friend of mine, after a visit to fix something for me, said ‘We’re very busy at the moment. And then some of our customers phone us up with problems, heh, heh.’ At the same time as saying this, he looked at me and winked, and both the wink and the laugh that accompanied and punctuated the utterance were supposed to let me know, I assume, that he both did and did not wish this remark to be allowed to stand. By using these signals, he was able to say that it was possible to take this as a joke and, at the same time, he managed to articulate a certain element of criticism, i.e. that I had asked him to do something, when he was in fact very busy. So FTAs are more complex than Brown and Levinson allow. Several theorists have criticised the notion of FTA, since the assessment of an utterance or act as constituting an FTA is the result of a process of judgement undertaken by speaker and hearer; their assessments of the seriousness or existence of an FTA is not something that can be clearly established by an analyst. Furthermore, as I show later in this chapter, in the analysis of example 1, a conversation between family members, M, P, and T, it could be argued that ‘in an atmosphere of empathy and respect partners are able to view misunderstandings as temporary breakdowns of communication rather than having to interpret them as threats to face’ (Janney and Arndt, 1992: 21). Thus, individuals have to decide whether an FTA has been committed or whether the possible infraction can be overlooked. Thus, I would like to take issue with the way that Brown and Levinson assume that it is possible to describe the constituents of politeness. If we assume that politeness is about judgements and assessments of utterances by participants, then the process whereby an analyst categorises an utterance as positive or negative politeness, or identifies an FTA, is rendered problematic.

80

Gender and Politeness

Problems with Brown and Levinson’s model of communication There are several difficulties with Brown and Levinson’s work on politeness which stem from their underlying model of communication. I should like to deal with the following difficulties in turn: their reliance on Speech Act theory; and their inability to describe politeness when it operates at the level of inference. Brown and Levinson’s allegiance to Grice, as I noted in chapter 1, leads to a problematic view of the way that politeness functions, as it assumes that politeness is an aberration, which can only be understood through a process of referring to a base speech act from which the utterance deviates, violating one of the maxims. For conventionalised politeness, such as ‘Could you possibly open the window’, when the elements ‘Could you possibly . . .’ first developed within English as a means of expressing a wish not to impose on someone, perhaps it was a strategy of mitigation as described by Brown and Levinson. However, that is not to say that present-day speakers go through the same process of deduction when they decide to choose this strategy rather than others when they want to ask someone to open the window, nor do hearers need to decode the utterance in this way. These strategies have become automatised and therefore do not need to be ‘worked out’ in the way that Brown and Levinson or Leech suggest. Jary (1998) argues that a great many politeness formulas go unnoticed by participants; it is only when they are stressed that they become salient to participants. He gives the example of a teacher saying to a class ‘Can you PLEASE be quiet?’ ‘stressing the please in order to make it manifest that she is behaving in accordance with the rules that govern their interaction and implicating that the pupils do the same. In such a case, one would probably want to say that politeness is linguistically communicated’ (Jary, 1998: 16). Furthermore, here the use of ‘please’ has more the function of a command than simply implicating politeness. Jary goes on to argue that one would not want therefore to see all instances of ‘please’ as being similarly communicative of politeness. He states that ‘the fact that stress is required to bring the choice of form to the attention of the hearers lends strong support to the claim that this choice generally goes unnoticed’ (Jary, 1998: 16). Brown and Levinson see politeness as a deviation from rational efficient communication, which they base on Grice’s Co-operative

Theorising politeness

81

Principle (CP): they state: ‘there is a working assumption by conversationalists of the rational and efficient nature of talk. It is against that assumption that polite ways of talking show up as deviations, requiring rational explanation on the part of the recipient, who finds in considerations of politeness reasons for the speaker’s apparent irrationality or inefficiency’ (Brown and Levinson, 1987: 4). However, this only holds if we assume that politeness is seen as a deviation each time that speakers encounter a polite utterance. In some senses, because some instances of politeness are largely routinized, an utterance such as ‘Could you hold this for me, please?’ is not, or not any longer, seen as a deviation – processing does not necessarily follow the route of viewing the utterance as a deviation from an idealised proposition of ‘Hold this for me.’ Perhaps this view of politeness as deviation can explain the development of certain types of speech style, but it cannot describe the type of processing that real speakers do when they hear an utterance which they are trying to decide whether to interpret as polite or not. A further problem with this view of politeness as aberration from a set of maxims is that the number of maxims is potentially unconstrained and the maxims themselves are based on relatively ill-defined terms (such as sympathy) which are insufficiently related to a linguistic function. What is omitted from accounts of politeness which focus on a turn-by-turn analysis of individual utterances or phrases is the higher level organisation of speech and discourse in general. Perhaps it is in this area that Brown and Levinson are prepared to acknowledge some of the shortcomings of their 1978 work, in that since they based their work on Speech Act theory they were led to analyse only short stretches of speech: ‘speech act theory forces a sentencebased, speaker-oriented mode of analysis, requiring attribution of speech act categories where our own thesis requires that utterances are often equivocal in force’ (Brown and Levinson, 1987: 10). They go on to argue that FTAs in fact need not be realised in sentences, but may occur over longer stretches of speech. However, in their analyses, the focus is generally at the level of the sentence only. They acknowledge this problem: ‘in our analysis so far, we have talked as if interaction were built out of unit acts, each of which might be an FTA requiring strategic adjustment of some sort or other, and which were strung together with no more than occasional reference to prior acts . . . or to succeeding acts’ (Brown and Levinson, 1978:

82

Gender and Politeness

237). Therefore, although they try to extend their work to examine adjacency pairs, there is a tendency for them to wish to examine longer stretches of speech, but to be prevented from doing so by their model of communication. They state: ‘FTAs do not necessarily inhere in single acts (and hence the concept might be better labelled “face threatening intention”)’ and ‘a high level intention to issue a criticism may be conveyed by a series of acts . . . that are not themselves FTAs . . . in short, plans – including conversational plans – are hierarchical, and conversational understanding is achieved by reconstruction of levels of intent beyond and above and integrative of those that lie behind particular utterances or sentences’ (Brown and Levinson, 1978: 238). They give an example of an interaction between an Indian woman who visits a member of her extended family; she describes to them in detail the expenses of a wedding in her immediate family and the other interactants respond by describing the expenses they have incurred due to an illness. It is clear to Brown and Levinson that this lengthy description of wedding expenses is a preamble to a request for money by the woman, although the request is in fact not articulated during the conversation explicitly. They note that this interaction continues for four hours. The visit by the woman is glossed by the family as ‘she came to ask for a loan’, but nowhere in the interaction is this stated explicitly by the woman. This is one of the examples where Brown and Levinson draw on a notion of what the interactants think is going on in a conversation and where they analyse longer stretches of speech; however, their model of politeness militates against longer stretches of speech being considered more generally.21 Rather than the sentence level, it is important to analyse at the discourse and the meta-discourse level. Taylor (1992) argues that we should distinguish between intellectual meta-discourse, that is theoretical analysis and thought; and practical meta-discourse, that is the thinking about what has been said previously in terms of the impact it makes on our relationships. This practical meta-discourse seems to be of importance for the analysis of politeness, because it is at this level, either in terms of the thinking about what other people mean in relation to their utterances now or the glossing that we give to statements made in the past, that assessments of politeness are made. For Taylor ‘an important function of meta-discourse is to serve as a means by which we may attempt to influence how discursive acts

Theorising politeness

83

and sequences (both our own and those of our interlocutors) are to be seen: that is, what aspect they are to be seen under’ (Taylor, 1992: 12). Thus, meta-discursive glossing is part of the process whereby we arrive at a judgement of whether polite or impolite acts have been uttered. Analysis of politeness which draws on the work of Brown and Levinson generally remains at the level of the linguistic utterance itself and does not analyse the meta-discourse level. Taylor says: ‘imagine I refer to what you just said as an insult . . . By this remark, I may succeed in influencing the rhetorical status which we give to your utterance in the remainder of our conversation; that is, whether we subsequently treat it as having been an insult . . . or something else’ (Taylor, 1992: 12). This lack of consideration for the very level at which assessments of politeness take place is a major failing in much politeness research. Culpeper has criticised Brown and Levinson’s model for being unable to analyse inference, which he suggests is the level at which a great deal of linguistic politeness and impoliteness occurs (Culpeper, 1996). As Holmes (1995) notes, politeness cannot be said to reside within linguistic forms.22 Thus, a statement such as ‘Do you think it would be possible for you to contact Jean Thomas today?’ would be interpreted by Brown and Levinson as polite, if used by a boss to her/his secretary, since mitigating features are included in this direct request. The request itself is unlikely to constitute an FTA because of the boss’s assessment of her/his position relative to the secretary (power: P) and the fact that the imposition is not particularly great (ranking of imposition: R) because it is the secretary’s role to perform such tasks. Thus, given Brown and Levinson’s framework, it is likely that this would be categorised as straightforwardly negative politeness, perhaps overly polite, given these assessments of P and R.23 However, this request might in fact be interpreted by the secretary as impolite, if it were said by a boss to his/her secretary in a sarcastic tone; if they usually have an informal style of communicating; and if the request had been made previously but the person had not been contacted. Without this contextual information about the participants’ assessment of their relationship (that is, generally informal, and therefore low social distance: D) in the light of the interactions which have taken place before this exchange, the analysis of the exchange would be incorrectly analysed as polite. In some ways, the surface politeness of the utterance may be masking an

84

Gender and Politeness

underlying message, which the hearer has to infer; over-politeness in relation to assumed norms of this particular community of practice may be being used here strategically to imply criticism, which may or may not constitute impoliteness, depending on whether the speaker and/or the hearer assume that that criticism is justified or not. It may well be the case that both participants would be very aware of the potential impoliteness in this utterance or they might disagree about whether it was polite or impolite; or they may concur in assessing it as slightly impolite (but presumably less impolite than ‘Why the hell haven’t you contacted Jean Thomas, when I asked you to?’), but have decided that it does not constitute an FTA, another possibility which Brown and Levinson’s model cannot account for.24 Thus, the very features which Brown and Levinson argue indicate politeness may in fact be used to express impoliteness and may also be interpreted as polite or impolite. Furthermore, as Diamond demonstrates, in certain contexts, those in positions of high status may well use language more normally associated with indirectness and mitigation than those in positions of lesser status (Diamond, 1996). She argues that in her analyses of a close-knit community of psychotherapists ‘the lower ranking members’ style of speaking approximates the higher ranking attributes of authority and power, while higher ranking members use strategies of solidarity, informal speech styles, personal and in-group markers’ (Diamond, 1996: 80). If this is so, then the assumption that it is possible to assess the relation between P and linguistic performance in any simple way must be questioned. To take another anecdotal example, a Libyan man was making a telephone call from a public telephone box and a British woman was waiting to make a call herself; when he finished his call, he had 80p left as a credit on the telephone. The woman, as he was leaving the telephone box, asked him if he could ‘lend’ her some money to make a call, and he told her that she could use the credit; in response, she said: ‘Thank you VERY MUCH; thank you VERY MUCH.’ He discussed this incident later with a Libyan friend because he felt troubled by it; rather then assuming that she had expressed gratitude for his generosity, he wondered whether in fact she had emphasised the words ‘very much’ to imply that he had been stupid to have been so generous to a stranger (Hamza, pers. comm). We can see this difficulty in the analysis of inference more clearly in an example from my data, where there are a number of inferences

Theorising politeness

85

which might possibly be made, some of which may be considered to be impolite by an analyst using a Brown and Levinson framework. It seems that the interactants here are in a constant process of trying to assess whether certain elements of the conversation could be interpreted as implying criticism and hence impoliteness. The interactants are a white, middle-class, professional woman, M (from a working-class background), and her two sons: the elder, P, in his thirties; and the younger, T, in his late twenties and at the time of the interaction at university. Within this particular community of practice – the men in relation to their mother – there seem to be particular distinctive discursive styles which none of them seems to adopt in any other environments: the elder brother plays a considerate, conciliatory role and the younger son plays a direct and potentially impolite role, both in relation to the mother and to his brother. The two brothers do not behave in this way when they are speaking together, nor do they adopt these roles with others to the same extent. This interaction takes place at the dinner table at a meal prepared by the mother: 1P: 2T: 3P 4M: 5P: 6T: 8M: 9T: 10M 11M 12M: 13T: 14T 15P: →

16M: 18T: 19M: 20P: 21T:

Do you want some of this wine? Do you want to have a (.) taste?= =Yeh I’ll have a bit. Maman. you too? Why not? Everybody else is(.) Ooops did you give them a dust? Yes Why are you not taking the sauce? Well [ ] Just like a bit of pasta on its own? [ ] No I’ve never had pasta on its own really just sort of reading recipes that sort of say(.) delicious with butter and black pepper and I thought to myself well I shall [ ] Shows how the priorities are important because I couldn’t AFFORD to eat anything else [ ] (laughs) [ ] Right has everybody got two sorts of everything? T?= =No Mm? (to T) T? It’s amazing how many meals you can get out of

86

Gender and Politeness 22T: 23M: 24T:

a 30p packet of spaghetti (laughs) [ ] Well I daresay. Do you put cheese?

Yeh I used to buy little packets of Parmesan= 25M: =I was going to say it should be fairly nourishing [ ] 26T: Is it HELL! [ ] 27M: What with cheese and [ ] 28T: with about a 29T: teaspoon of Parmesan and the rest of it’s just carbohydrate [ ] 30M: well why why well why didn’t you er grate Cheddar over it? 31T: Because I couldn’t afford any that’s 32T: why= 33M: =Well I would have thought a chunk of Cheddar wouldn’t have been 34M much more expensive than a tiddly bit of Parmesan [ ] 35T: About 25p for a carton of Parmesan it’ll 36T last me for weeks(.) can’t even SEE 25p’s worth of Cheddar= → 37M: =Pass the salad please T. Well your very good health children. Cold for → 38M breakfast tomorrow morning. I don’t think I’ve ever had any that’s four 39M years old before. 40P: It hasn’t deteriorated at all. It’s lovely (.) different to usual (Data: 19.1) EXAMPLE 1

The discussion about eating pasta seems to be being used by T to bring the subject round to the fact that he has not enough money to provide adequate food for himself while he is at university – a topic which occurs in other conversations. Thus, there may be an inferred criticism of his mother here in terms of the amount of money that she gives him to live on. Thus, the discussion about whether Parmesan or Cheddar is more economical may be a veiled discussion of conflicts over money: whether enough money is provided; whether the money that is provided is spent wisely enough. The tension in the air at this moment in the discussion is palpable, since neither T nor M will drop the topic of conversation which seems to be indirectly and problematically alluding to financial problems, even though M tries to change the topic at line 16 by taking the topic back to whether everyone has helped themselves to enough food:

Theorising politeness →

87

14T

because I couldn’t AFFORD to eat anything else [ ] 15P: (laughs) [ 16M: Right has everybody got two sorts of everything?(.) T?=

T refuses to respond to this, and she is forced to ask him explicitly by name to engage in a conversation about whether he has had ‘two sorts of everything’ rather than a conversation on food at university. In this section, both P and M refuse to take up the implications of T’s implied criticism at line 14, about his having to eat pasta on its own, in contrast to his mother who has cooked it from choice. At lines 19 and 20, when both P and M try to get T to respond to them by explicitly questioning him, at first he simply does not respond, and then does not respond to their question about whether he has ‘two sorts of everything’ to eat, but instead continues to detail how little he can spend on food while he is at university, although he does laugh at the end of this utterance: 16M:

→ →

18T: 19M: 20P: 21T: 22T:

Right has everybody got two sorts of everything? T?= =No Mm? (to T) T? It’s amazing how many meals you can get out of a 30p packet of spaghetti (laughs)

M tries to move the topic of conversation onto the topic of the current meal and the wine again, at line 37, a topic which P takes up.

→ →

35T: 36T 37M: 38M 39M 40P: 41P

About 25p for a carton of Parmesan it’ll last me for weeks(.) can’t even SEE 25p’s worth of Cheddar= =Pass the salad please T(.)Well your very good health children. Cold for breakfast tomorrow morning. I don’t think I’ve ever had any that’s four years old before. It hasn’t deteriorated at all. It’s lovely (.) different to usual

M consistently tries to change the topic of conversation quite forcefully here; this is not simply self-effacement and negative politeness, but rather it seems to be a strategic decision not to take up the implications of T’s utterances. However, although there are clearly

88

Gender and Politeness

tensions about the inferences which can be made from this discussion of food here, it is not clear to me from this extract that this indirectness on the part of P and M particularly and perhaps also on the part of T can be easily classified as politeness (as he is not openly criticising his mother), or that it is a simple matter of working out what inferences are being made and accepted as being ‘in play’ by all of the participants. Each participant has the option of taking up the inference and making the implied criticism explicit, but none of them does this. But can this leaving of topics at the level of inference and refusing to taking up inferencing rather than making them explicit be considered as necessarily polite, in the strategic individualistic way that Brown and Levinson do? It is debatable whether this exchange is considered by any of the participants as polite since it was so tense, as P and M remarked to me when questioned about it afterwards, but it is also debatable whether this exchange could constitute an FTA on the part of T, for not responding promptly or for continuing to talk on a topic when others have tried to change the topic, or on the part of M, for refusing to acknowledge T’s implied criticism and for changing the topic to the organisation of food at the table. Although the discussion was difficult and tense, as subsequent interactions proved sometimes to be, it is not clear that T’s interventions are glossed by either M or P as being ‘impolite’. Rather perhaps they could be seen as constituting a threat not to the face of M but to the smooth running of the interaction as a whole, and therefore to the community of practice within which this interaction takes place. T’s comments are also seen by the other participants less to be an FTA than to be considered as part of his interactional style within this context, and perhaps also a product of his reaction to the particular financial and physical circumstances that he finds himself in. It is unlikely that T himself would gloss his utterances as impolite, since he characterises himself generally as direct and plain-speaking.25 As Wodak remarks: ‘there exists not one discourse . . . but a whole set of interwoven, conflicting discourses which construct and establish multiple relationships’ (Wodak, 1996: 12). Each interactant here is establishing and affirming/challenging particular positions for themselves within the family, at a time when those roles are being renegotiated because of changing circumstances – one son working away from home and another having recently left home. T at least is still dependent on his mother

Theorising politeness

89

in part and yet independent in that he lives away from home and is adult. There is thus a range of different discursive threads which are at work within this interaction, which may be interpreted in a range of different ways. Within families, this sort of give and take over inferences which may or may not be being made and which may or may not be acknowledged by the group to be ‘in play’ cannot simply or easily be glossed as constituting polite or impolite behaviour. By changing the topic here, M shows that it is possible skilfully to avoid a situation where an FTA might be committed, where open criticism might be made and the smooth running of the interaction and the evening might be threatened. These problems in Brown and Levinson’s model are due to a model of communication which tends to assume a simple code-like transmission of information between participants. As I suggested in chapter 1, the model of communication that I propose is much more based on the notion that in conversation we communicate, both explicitly and through implication, a wide range of types of information both about ourselves and about our relations to others, and in the process of communicating we enact those relations. Because of the complexity of designating what exactly it is that we are communicating, this information is often not clearly understood by others. Problems with Brown’s and Levinson’s methodology At a methodological level, there are also problems with Brown and Levinson’s work; here I discuss the difficulties entailed in their datacollection, interpretation, their analysis of variables, and the claims they make for their model. Brown and Levinson describe the speaker and hearer using a variety of models, but the main model of the speaker which informs their work is that of the Model Speaker described in the previous chapter. Paradoxically, they only analyse the hearer when s/he becomes a speaker and not when s/he is actually processing what the speaker is saying. Thus their analysis is of speakers alone. They analyse invented examples between two imaginary speakers, Mary and Peter, alongside transcribed data from several different language groups. This mixture of data is something which they themselves problematise in their later work, and it is something which poses

90

Gender and Politeness

problems for their claims about the universality of politeness. Invented examples between idealised speakers seem to fit the model better than real data from conversations. As I argued in chapter 1, simply analysing transcribed data is not sufficient to capture the complexity of politeness: Janney and Arndt state that ‘modifications of verbal directness and intensity, variations of intensity, variations of intonation and tone of voice, changing facial expressions, shifting glances and other activities provide a running commentary on what is said literally’, and these elements play a major role in people signalling that something is to be taken as polite, and their interlocutors accepting or rejecting an utterance as polite (Janney and Arndt, 1992: 35). Brown and Levinson’s model can further be criticised for the way that they interpret their data, since they assume that it is possible to know what a polite or impolite act means. It is thus a model of interaction which is focused on production, i.e. which conflates the intentions, or the perceived intentions, of the speaker with that of the meaning of the interaction as a whole. As such, it assumes that the analyst can simply work out the intentions of the speaker. Because of their reliance on the notion of a Model Speaker, Brown and Levinson assume that speakers have clear intentions which can be unproblematically decoded by the hearers: ‘intentions of actors are reconstructable by observers of recipients of actions’ (Brown and Levinson, 1987: 7). They themselves note several challenges to this notion: conceptual impossibility, psychological implausibility, and cultural bias. The conceptual problem relates to the problem of infinite regress that I mentioned in chapter 1 in relation to the dialogic nature of assumptions about other speakers’ intentions and the construction of one’s own intentions in relation to one’s perception of theirs. Brown and Levinson argue that their earlier notion of mutual knowledge, which they later revise to Sperber and Wilson’s notion of mutual manifestness, solves this problem. As I mentioned in the previous chapter, the problem with using idealised Model Speakers is that one is forced to assume that the speaker and hearer share knowledge or at least those elements which are mutually manifest; however, Brown and Levinson state that they are aware of the problem of this type of assumption since ‘assessments like . . . whether an actor is known to enjoy being imposed on . . . raise a very complex problem, that of assessing the status of mutual knowledge in a given

Theorising politeness

91

interaction. How do we know what is mutually known, and how do we know we know’ (Brown and Levinson, 1978: 297). This problem cannot be resolved if we adopt the notion of the Model Person/ Speaker. We cannot know whether our intended politeness is perceived as such by others and is not perceived by them as impolite or non-polite.26 As I mentioned in chapter 1, there is a general problem in Brown and Levinson’s model, with the notion of intentionality, related to the question of agency. Particularly with politeness, as I have shown above, there are verbal habits which add up to the parameters of our style within particular interactions, and the situation itself makes certain demands upon speakers and hearers in terms of what is considered by us and others to be appropriate in politeness terms. Thus, what is complex about politeness is that the hearer constantly has to assess whether the politeness which s/he has judged to have been expressed in the other’s speech, is due to the demands of the situation or is in fact a result of a decision to be polite by the speaker. Thus, rather than assuming that all politeness is intentional and strategic, when, for example, you want to borrow a pen from a stranger in a Post Office (if you can see that he/she has a pen) you might say ‘Can I/Could I borrow your pen for a minute?’ because the context, and the history of such interactions in English, generally demands it. These social factors and context cannot be considered if we rely on a notion of politeness as decided upon by the individual alone; as Werkhofer argues: ‘The antagonism between the individual and the social is not only reflected by the premise that the intention of the speaker is a face-threatening one, the whole mechanism of generating an utterance is characterised in mentalistic, intentionalistic and linear terms. As it is triggered by an intention of the speaker, social factors can only come in later’ (Werkhofer, 1992: 180). I would argue that a large part of politeness is judged by speakers to be necessary because of social constraints or ideologies/discourses which speakers have internalised as their own values, but which in fact are those hypothesised of the dominant group. As I argued in the previous chapter, the Model Speaker generally drawn upon in linguistic analysis draws on a form of reasoning which Brown and Levinson term ‘practical rationality’: ‘a precisely definable mode of reasoning from ends to the means that will achieve those ends’ (Brown and Levinson, 1978: 63). Practical rationality

92

Gender and Politeness

is a form of cost-benefit analysis where politeness is perceived as a type of symbolic recompense for some debt to the other; for example, the act of thanking someone ‘repays’ that person symbolically (Haverkate, 1988). Indeed, Sperber and Wilson’s (1986) model of cognitive processing in terms of relevance is, in essence, a costbenefit analysis. This form of reasoning works well when we analyse invented examples, but with transcribed data we can see that interactants often work with a number of different short-term and longterm plans, some of which are shared with others and some of which are apparent only at certain stages in the interaction. Interactants try to make sense of what is happening, as Toolan (1996) argues, but whether that process of making sense can be described under the heading of practical rationality is debatable. Consider the following episode between a white, middle-class male director of studies of a foreign language teaching centre (T) and a white, middle-class female teacher (F); F has been called into T’s office. 1T: 2F: 3T: 4F: 5F 6T: 7F:



8T: 9T: 10T: 11F: 12T: 13F: 14T: 15T: 16T: 17F: 18T: 19F: 20T:



21F: 22T:

Now I sent you a little note about (.) arrangements for the conference yeh?= =Yeh right What I thought of offering you or perhaps I don’t know [ ] [ yeh well(.) look did you send me a note about it?= =Yeh didn’t you get it?= =I saw the main note about uh you know [ ] yeh yes um that was the full sheet= =the reasons of areas yeh no I don’t think I did but G mentioned it ah I see well I sent you a little note saying I wanted to talk to you about that week because (.) especially as you would normally have been teaching uh you know and assuming that we can get substitutes [ ] yeh for you= =mmm= =which is proving a bit of a problem I thought it would [ Well er J has been [be

Theorising politeness

93

23T:

appropriate(.) has she? uhuh great. Interested? [ 24F: yeah but= 25T: =yeah but what? → 26F: (laughs) I don’t know 27T: [Why?] 28F: I’m just very conscious 29F that teachers represent the organisation that’s all= 30T: =yeah 31F: you know(.) yes(.) but on the other hand(.) er. better to have 32F: someone who knows the situation for a week I mean if she’s done it [ ] 33F: well that’s true 34T: for five weeks= 35F: =yes= 36T: =you know I mean we can’t say that she’s er not appropriate for one 37T: week 38F: yes 39T: really= 40F: =yes 41T: ANYWAY that’s at least on the way to being sorted out isn’t it? Data: 92.1 EXAMPLE 2

This extract illustrates the practical difficulties in analysing politeness using a Gricean model of conversation and an idealised Model Speaker as Brown and Levinson have done. In a sense, we as analysts will never know exactly what is going on in this conversation, because we do not know what happened before this interaction: that is the interactants’ interactional history. There may be elements of the conversation which have a specific meaning or difficulty for the interactants, because of previous conversations with each other or with other members of the workforce at the language centre, for example, the reference to J from line 21 onwards is opaque to me in relation to the topic of conversation here, but is clearly not to the interactants. There may be particular issues which have especial resonances for these two speakers, but which are unclear to us as analysts. What we can notice about this extract is that T appears to find it difficult to approach the subject of money which he is trying to bring up; this can be seen from the fact that he takes from line 1 to line 14 to bring the conversation around to the ‘little’ note which he has sent to F, in which he suggests that she might like to represent the company at a conference. What seems to be causing the difficulty is that T needs to negotiate with F about the amount of money T will be paying F in addition to her salary to represent

94

Gender and Politeness

the organisation at the conference (which he later goes on to do: see example 3), and from the rest of the conversation, it is clear that this is what he has asked F into his office to discuss: exactly how much money F would settle for. That this is clear to both participants, even though the word money has not been uttered in this extract, emerges at lines 24–27, where F states that she is interested, but expresses some reservations. She does not complete her sentence, even when T asks her ‘Yeh, but what?’ 23T:



24F: 25T: 26F: 27T:

appropriate(.) has she? uhuh great. Interested? [ yeah but= =yeah but what? (laughs) I don’t know [Why?]

Here, she laughs and states that she does not know, which we may take literally, or as an expression of her discomfort in the conversation, since money has not been mentioned and it is difficult for her as an employee to initiate a discussion of money (line 26). It is unclear whether her saying ‘I don’t know’ and also drawing attention to the fact that she would be representing the organisation, even whilst she poses this as a general and not specific statement about teachers and not herself: 28F: 29F

I’m just very conscious that teachers represent the organisation that’s all=

is motivated by an implicit assumption that she should be receiving a wage commensurate with the level of responsibility of the position; or whether she is, in fact, asserting that because it involves some responsibility she does not want the position; or whether she is referring to herself here or to the other person, J, who is mentioned in line 21ff. (since no pronouns are used). F may also not have realised that T had called her in to negotiate over money at all, but had thought it was to simply ask her to agree to attend the conference. If we analyse this conversation using only a model of practical rationality and strategic use of politeness on the part of both interactants, we are unable to capture the way in which both parties are constrained by the difficulty of negotiating about money, especially when one of them is in a position of relative power and status in relation to the other. That is not to say that F does not

Theorising politeness

95

have some interactional power, since it is open to her not to state explicitly what it is she wants, as she does in line 26, by stating ‘I don’t know’ rather than, for example, stating clearly that talking about this is difficult, or stating that she does not understand. Thus, both participants seem in this extract to be able to operate on a certain level of trying to make sense of what is going on, trying to get the gist, hypothesising about the possible meanings that are being hinted at, rather than having to understand fully the import of each individual word or move. The conversation continues for another two minutes (quite a long time for such an interaction) in a similarly rambling and inconclusive fashion, before T broaches the subject of money at all:

1T: 2T: 3T:

We’ll play it that way(.) um for(.) F(.) what I thought was(.) as it’s a week that you (.) would normally be working(.) I feel we ought to offer you (3) something for the work that you’d be doing for us(.) I mean [ ] 4F: mmm 5T: be down there. just generally as a conference participant= 6F: =right= 7T: =but you won’t be totally free(.) because they’ll set up this afternoon 8T: what [ ] → 9F: will I have time to buy a compass? [ ] 10T: your sort of duties are(.) you will yeh you’ll have 11F: [(laughs)] 12T: time to buy a compass(.) that could be time-tabled but I thought of 13F [right] 14T: offering (.) now how much did I think of offering(.) forgotten the figure 15T: (.) dear me I might say er say the wrong thing here [ → 16F: Would you like to 17F: adjourn for six seconds?= 18T: =No five hundred for the week 19F: yeh 20T: How do you feel about that? That’s 21T: on top of your subsistence(.) yeh? [ ] 22F: (laughs) do I have to say now?= 23T: =well= 24F =that’s OK th that’s OK [ 25T: yeh is that alright?= 26F: yeh OK Data: 92.2/3 EXAMPLE 3

96

Gender and Politeness

In a sense, what is missing from Brown and Levinson’s model of politeness in conversation is just how much of what goes on is coconstructed by the participants. Because Brown and Levinson tend to focus on speaker intentions and practical rationality, they do not analyse the way that the intransigence or co-operation of the person we are speaking to determines our assessment of the need for a certain level or type of politeness. Here, F could be interpreted as not co-operating with T: when he cannot remember the figure that he is about to offer her (the very reason that we assume she has been called into the office): 12T: 13F 14T: 15T: →

16F: 17F: 18T: 19F:

time to buy a compass(.) that could be time-tabled but I thought of [right] offering (.) now how much did I think of offering(.) forgotten the figure (.) dear me I might say er say the wrong thing here [ Would you like to adjourn for six seconds?= =No five hundred for the week yeh

F offers in line 16, politely, or perhaps strategically, manipulatively, ironically, or even sarcastically – it is not evident which – to adjourn the meeting for six seconds. Again, she does not immediately agree to T’s offer of £500 at line 22, but at first seems to be asking for more time to consider (possibly stalling to bargain for more?) and even then does not wholeheartedly agree to the offer until line 26. 21T: →

22F: 23T: 24F 25T: 26F:

on top of your subsistence(.) yeh? [ ] (laughs) do I have to say now?= =well= =that’s OK th that’s OK [ yeh is that alright?= yeh OK

Thus, rather than viewing this interaction as a ‘product’, and seeing the conversation as a finished text, with interactants formulating their intentions before the conversation, if we use a more processoriented model of analysis, we can see that this conversation could have developed in a number of very different ways. There is a negotiation here over what face work is done. It is not clear to me that either of these interactants is clearly using positive politeness

Theorising politeness

97

or negative politeness only, in any simple way, nor that the status of T in relation to F is demanding a particular style of politeness from either T or F. In Brown and Levinson’s model, the fact that T is in a more powerful institutional role in relation to F should mean that F gives more negative politeness to him, but it is difficult to argue that this is the case here; indeed, the false starts and hesitations could be argued to be almost entirely on T’s side and it is difficult to assume that these hesitations and mitigations are interpreted as politeness and deference by F. They may simply signal a difficulty with knowing what to say in this particular tense situation, or they may indeed be part of T’s verbal style, despite the fact that he is the director of a school. Thus, rather than seeing conversation being carried out by speakers with clear intentions about what they want to achieve, I would argue that those intentions are often only formulated in the interaction itself, or that long-term aims are shaped by the short-term progress of the conversation itself. Werkhofer argues that Brown and Levinson assume linear processing on the part of the speaker, that is, the intentions of the speaker follow in a simple linear pattern (first, second, third) whereas processing ‘may be of a “cyclical” or “parallel” type, that is the speaker may, while s/he is already speaking, go back to the planning stage, and s/he may do so repeatedly, thus continuing to make up and change his/her mind on what to say next, depending perhaps on the changing impressions of how the situation at hand develops and on how well s/he feels s/he has been faring thus far’ (Werkhofer, 1992: 169). This seems to be the case with examples 2 and 3 above where F and T are testing out what each of them can achieve, depending on how assertive or demanding the other is perceived to be. T uses reformulations, that is, a summing up of the gist of what has gone on before, which Thornborrow (2002) has argued are more likely to be used by those in positions of higher status, for example, T’s summing up move in example 2, which tries both to achieve and assume that something is completed at the same time: 41T:

ANYWAY that’s at least on the way to being sorted out isn’t it?

and his move in example 3: 1T:

We’ll play it that way(.) um for(.) F(.) what I thought was(.) as it’s a

98

Gender and Politeness

where he sums up the preceding segment of talk by saying ‘We’ll play it that way’ before moving on to discuss the question of money. However, whilst F does not use reformulations, that is not to say that F does not also use strategies and moves which could be seen as characterising powerful speech, for as I have noted she uses quite ambiguous utterances, which force T to spell out the terms of the negotiations. They interrupt each other more or less equally. In this instance, I would argue that it is only individuals interacting within particular communities of practice who will be able to assess whether a particular act is polite or impolite, and even then, such interpretations may be the subject of disagreement and misunderstanding. Furthermore, there may be instances where coconversationalists simply do not know whether an utterance is to be taken as polite or impolite, and whether this presumed politeness or impoliteness may or may not be significant. Most people cope very well with not knowing exactly what is going on in a conversation, and politeness may well be a form of linguistic behaviour, or a linguistic resource, which may help that process of simply getting through a conversation: for example, a school bus driver (A), whom I (S) speak to every day for approximately the time it takes for my children to get on the school bus, said to me one morning: A: S: A: S:

Hello Mum, Chief Seattle’s address. Got it off the Internet. Nature and respect Hmm, it’s good, isn’t it. The kids should read it every day Yes, Have you got the site address?

In this interaction, I, at first, had no idea what the person was talking about – I was taken aback by being addressed as ‘Mum’ and I did not immediately realise who he was referring to when he said ‘Chief Seattle’, as this was not prompted by a previous reference or conversation, nor did I work out which way he was using ‘address’, whether as an actual address (Internet or otherwise) or in terms of a speech. Thus, since I assumed from his facial expression, his intonation, and from the ending of his utterance with ‘nature and respect’, both of which are generally positively inflected terms, that I was required to respond positively, my response was simply a hazard at recycling what he had just said in the broadest of possible terms. This could be seen as a polite and thus self-defensive move on my

Theorising politeness

99

part, so that the conversation did not break down, which might have happened if I had simply said ‘I have no idea what you are talking about.’ Similarly my asking for the site address was not a real inquiry but simply to show that I was keeping up with the conversation. I was ‘just being polite’. Thus, although I have no idea how the school bus driver interpreted my responses, I felt that I was signalling a polite wish to engage in conversation, although I had little idea of the import or relevance of what was being said to me. As I discussed in the previous chapter, it is not possible to say whether the bus driver interpreted what I said as polite, incomprehensible, or patronising. As interactants, however, we are broadly tolerant of a great deal of not entirely comprehensible discourse. Thus, Brown and Levinson’s model can be seen as an idealised model of perceptions of speaker intentions, but it cannot deal with the complexity of actual interactions where hypotheses about speaker intention and meaning constitute a much more complex, negotiated phenomenon.27 When discussing politeness, Brown and Levinson and other theorists following their model of analysis analyse only utterances and cannot attempt to analyse silence, even though they stress the importance of ‘off-record’ as a politeness strategy. As many theorists, such as Tannen and Saville-Troike (1985) and Tannen (1990), have shown, silence in fact is often a key feature in politeness and impoliteness. To give a banal example, an older, white, female upper middle-class friend said once when explaining why they had gone out shooting the previous day (something to which she knows I am opposed): ‘Well, of course, we are a shooting family.’ This seemed to me a ludicrous statement – the notion that one produces oneself as a type of family and justifies one’s actions accordingly, particularly this type of self-aggrandising aristocratic family position – but because I am fond of this person, I suppressed the cutting remark or laughter which I felt, out of concern for her, and because in my opinion our relationship could not stand such criticism. I think this is a quality of a great number of relationships with others, that there are certain opinions or information that we leave unsaid or that we actively suppress, because of our consideration for the other person’s feelings or because strategically we know it would not be in our interests.28 This suppression of utterances can be labelled offrecord within the Brown and Levinson model, but in essence it is unanalysable, as what can be analysed in their model is that which

100

Gender and Politeness

is representable. As an analyst, it is clear that we simply cannot access this level of motivation; however, we need to be aware that it plays a major role in politeness behaviour, and that if we use Brown and Levinson’s model we are restricting ourselves to the analysis of politeness which is representable. Brown and Levinson consider a number of variables which might affect the level of politeness that is employed by particular speakers; they sum up these variables under three headings: Power (P); Distance (D); and Rank of imposition (R). The model of power that they use is premissed on the ability of one person to impose their will on another, what Foucault terms the ‘repressive hypothesis’, that is a model of power which is about denying freedom to another (Foucault, 1978). However, power is much more complex than simple repression alone, and Foucault argues that power is productive as well as repressive, as I suggested in chapter 1. Brown and Levinson suggest that there is a simple correlation between power and politeness: ‘as S’s power over H increases, the weightiness of the FTA diminishes’ (Brown and Levinson, 1978: 83). However, the assessment of one’s power in any situation is something which is dependent on a range of factors: one may have an institutional role which may accord you certain privileges, but your power is not simply a matter of your position in an institution. It is something which is negotiated throughout all conversations and which is never finally achieved, much in the same way as Butler describes the performance of gender identity (Butler, 1993). Diamond distinguishes between ‘institutional status’ and ‘local rank’: ‘those social variables whose meaning is internal to a particular community’ (Diamond, 1996: 10). It is this level of bargaining over power relations which is most ‘up for grabs’ and which I refer to as interactional power. This is clear in examples 2 and 3 analysed above, where T’s position is one of superior institutional status, but the local rank or interactional power which he manages to negotiate for himself is constrained by F and by his assessment of the difficulty of the situation. We might also consider, for example, the way that in many institutions those who are officially assigned the role of making the decisions about the forward planning of the company are not necessarily the same as those who actually make the decisions. Thus, someone may have a ‘powerful’ position in the company, but employees soon learn that if they want to get something done, they need to approach someone

Theorising politeness

101

else who holds the real power in the company. Those in positions of institutional power may be perceived to be ineffective in their job. Diamond argues that we consider people to have achieved high local rank if they are able to win an argument, introduce a new topic, bring about a reform, change existing structures, lead a discussion, hold sway over others, and she goes on to say that ‘by defining power as political effectiveness we are stressing the fact that power is not merely a quality that is assigned or earned; it is also an interactional skill and process’ (Diamond, 1996: 12, emphasis in original). What is interesting about this description of negotiations of local rank and power is that it is primarily a description of linguistic processes. One might alter one’s level of politeness and deference to these people accordingly. Some of the people whom I interviewed remarked that they were less deferent and respectful to those in positions of power whom they regarded as ‘wankers’; they clearly distinguished between the institutional status of these people and the interactional power that they were accorded. Institutional power obviously plays a major role in the assessment of what behaviour is appropriate in a particular context, but assessments of effectiveness and competence, negotiations with those roles, and other factors play an equally important part. Social distance (D) is the degree to which interactants are familiar with one another: Brown and Levinson describe D in the following terms: ‘frequency of interaction and the kinds of material and nonmaterial goods (including face) exchanged between S and H’ (Brown and Levinson, 1978: 82). However, social distance cannot be characterised as achieved or stable. Brown and Levinson assume a stability in relationships, so that interactants are consistently distant or familiar and are both in agreement as to the degree of familiarity that it is appropriate for them to use. I would argue that social distance, because it is, like power, not something which is ever discussed explicitly, but which is negotiated in each interaction, is a variable about which interactants might have different perceptions. Fukushima remarks that ‘one of the difficulties [in assessing social distance] is due to the fact that the relationships among speakers are dynamic and open to negotiation’ (Fukushima, 2000: 83). The rank of imposition (R) is again a matter for negotiation, because it is not clear that interactants always agree on their perceptions of how much of an imposition a particular request is.

102

Gender and Politeness

Brown and Levinson’s view of the variables which have an impact on the level of politeness leads them to assume that these variables are both mutually manifest to all participants and also equally salient for all. But consider Jary’s comment: ‘should one party employ a form or strategy which does not fulfil these expectations [of P, D, and R], the result will be a change in the addressee’s cognitive environment. Moreover a form or strategy of this type would be highly relevant to the hearer as it would constitute evidence that the speaker ranked one of the three variables in a manner incompatible with the hearer’s assumptions about their mutual cognitive environment’ (Jary, 1998: 5). Furthermore, Brown and Levinson assume that each of the participants is clear about their social standing in relation to each other; in a situation where the participants are unsure about their relation to each other in terms of status, power, and social distance, Jary argues that: ‘in such cases, the forms used and strategies employed will serve as indicators of the speaker’s estimation of her status in relation to the hearer’ (Jary, 1998: 17). Thus, those elements of speech which are seen by participants to be polite indicators will be interpreted not simply as indicating concern for the others’ face needs, but will serve to indicate the estimation of status and social distance that each of the participants has so far made. Thus, rather than these variables determining the production of politeness behaviour, we can argue that so-called polite behaviour plays a major role in interactants working out their role and status in relation to each other. Not only do Brown and Levinson sum up a complex array of variables under these three headings, but they also say that they can compute the level of politeness from formulas that they have developed. For example, they state that power, social distance, and rank are the three factors which lead to a judgement of the weightiness of the imposition, leading to the choice of high or low politeness strategies: Wx = D(S, H) + P(H, S) + Rx Here the weightiness of the imposition [W] and thus the level of politeness strategy, can be computed from ‘adding together’ the social distance [D] of the speaker and hearer, their relative power [P], and the rank of the imposition [R]. But we might ask ourselves what exactly this process of adding these elements together means. In real terms, how easy is it to compute the power relation or the

Theorising politeness

103

social distance between two people except perhaps in terms of high and low? These formulas seem so schematic that they do not relate to language production or interpretation in any meaningful way. Werkhofer argues that ‘the weightiness of the imposition implied by an utterance does not seem to determine the degree of politeness that is employed’ (Werkhofer, 1992: 171). Brown and Levinson qualify their assertions about these variables by saying: ‘we are interested in D, P, and R only to the extent that actors think it is mutual knowledge between them that these variables have some particular values’ and I would agree that it is the interactants’ assessment of their relative power or distance which is important for politeness (Brown and Levinson, 1978: 79). However, it is unclear how, within Brown and Levinson’s model, it is possible to determine the interactants’ assessments, because the only assessments which are given are those of the analyst. Furthermore, there is no scope within their model for mismatches or disagreements between speakers and hearers about assessments of power and social distance. Watts et al. argue that these formulas are problematic: ‘Surely, the degree to which a social act is considered to be an imposition . . . depends crucially on P and D. So in order for the model person to be able to assess the value for R, s/he has to be able to calculate values for D and P first. No indication is given as to how this might be done’ (Watts et al., 1992: 19). Furthermore, there is no indication of when in the course of the conversation such an assessment is made, since assessments may alter during and indeed after an interaction. With this simplified view of the variables which affect the production of utterances, there are many factors that are not considered. Just to give one example, Brown and Levinson do not consider the variable of age, and yet there seems to be a considerable difference in utterances, and judgements about utterances, made by people who are older (Scollon and Scollon, 1995). In the interviews that I conducted, there is a strong perception among older people that young people are not as polite as they used to be, but this may be due largely to stereotypical beliefs that older people may have of younger people which is represented through their assessment of politeness. Younger people in the interviews tended, though not unequivocally, to regard politeness as irrelevant or trivial, when compared to the importance it is accorded by older people.29 Furthermore, Brown and Levinson are aware that in an utterance the variable which is most salient for the interaction may not

104

Gender and Politeness

be agreed upon by both participants. The mood of the participants may affect the way that these variables are assessed, for example, as they mention, on some days we may see everyone as a friend and on the next, we may see everyone as a potential enemy: Brown and Levinson state that ‘in order for interactants to interpret utterances correctly they must have some assessment of each other’s current mood’ (Brown and Levinson, 1978: 237). But, if this is so, they give no indication of how we as analysts can factor this into their model (which is, in fact, predicated on excluding such factors). In addition, as Brown and Levinson note, politeness is intricately related to whether a member of a group likes or dislikes another member of the group. Depending on the degree to which affect is already established, politeness will be judged accordingly; if someone is disliked, behaviour which in other contexts with different participants would be judged neutrally or as polite, will be judged to be impolite. Politeness itself is used as an ongoing weapon or justificatory principle in questions of whether someone is liked or disliked. Brown and Levinson state, in the second edition of their work, that they need to distinguish between affect and distance/familiarity: ‘formality (and other sorts of situation and setting classifications . . . will have a principled effect on assessments of FTA danger’ (Brown and Levinson, 1987: 16). However, simply adding more variables without a notion of how to measure these assessments of affect or formality does not help the assessment of politeness. Brown and Levinson themselves seem to suggest that quantitative analyses using their strategy are not the best way to proceed: ‘politeness is implicated by the semantic structure of the whole utterance, not communicated by “markers” or “mitigators” in a simple signalling fashion which can be quantified . . . In our view quantitative evaluations of polite redress in natural language data must always be preceded by and supplemented with qualitative ones’ (Brown and Levinson, 1987: 22, emphasis in original). However, this mixture of quantitative and qualitative methods cannot operate effectively in the analysis of politeness if what needs to be measured is the assessment of politeness and the factors which contribute to that assessment. Instead, we need to add to this qualitative analysis of data an examination of the judgements of interactants. A final difficulty at a methodological level is the claims that Brown and Levinson make for their model. They assert that politeness itself

Theorising politeness

105

is universal and that it functions in much the same way in all languages, with slight differences of emphasis. They also claim that their model is universally applicable, justifying this claim by drawing on data from three different languages. This claim is the one which poses most difficulty both at a theoretical and a methodological level. Many critics have argued that their claim of universality is unfounded, or based on Eurocentrism, extending a model of language which works at some level for English to all other languages (Bargiela et al., 2002). Even here, it is largely the speech of the dominant class which is taken as the exemplar, and the model does not sufficiently address the different emphases in other language groups (Mao, 1994; Boz, forthcoming). They themselves state that: ‘it may be thought that our universalistic account is an inexcusable cultural denudation or worse ethnocentric projection’ (Brown and Levinson, 1987: 13). However, they argue that despite these perceived difficulties with the emphasis that different cultures have in relation to face and the degree to which certain cultures tend to positive or negative politeness, they consider that their model still holds. Brown and Levinson claim that cultures, as a whole, tend to certain styles of politeness: the ‘ethos’ of a culture being ‘the affective quality of interaction characteristic of members of a society’ (Brown and Levinson, 1978: 248). Those cultures which they claim are primarily positive politeness cultures, for example America, will in general view the weightiness of imposition as of low importance; whereas those they see as negative politeness cultures, like Japan and Britain, will see impositions as something to be handled with great care and tact. However, Sifianou argues that we should not accept this characterisation of China and Japan as negative politeness cultures, for she argues that it may be the case that ‘concepts of “deference” and “formality” in oriental cultures differ from those in Western cultures’ (Sifianou, 1992: 211). Thus, when we analyse deference in Asian cultures, we may be working with a Western model of deference which does not fit those cultures; this behaviour may not be seen as deferent as such by Asians themselves. Brown and Levinson go on to extend this cross-cultural analysis to classes and dominated groups: ‘subcultural differences can be captured: . . . dominated groups . . . have positive politeness cultures; dominating groups have negative politeness cultures. That is, the world of the upper and middle groups is constructed in a stern and cold architecture

106

Gender and Politeness

of social distance, asymmetry and resentment of impositions, while the world of the lower groups is built on social closeness, symmetrical solidarity and reciprocity’ (Brown and Levinson, 1978: 250). However, Brown and Levinson seem to be confusing several very different contexts of speech. Perhaps what is captured here is the ‘public discourse’ of the upper classes, that is the type of speech used by the upper classes to others in public. The type of language which they characterise ‘lower groups’ as using is in fact that of the ‘hidden transcripts’, that is the type of language which is used by ‘lower groups’ when they are among their own kind. This may differ markedly from the ‘public discourse’ of working-class people when talking to those from other classes (Scott, 1990, see fn. 28). This tendency to characterise classes and cultures as homogeneous is not easily sustained when we examine the complexity of politeness in even one culture, or even within one class, and seems to be dependent on stereotypical beliefs about the linguistic behaviour of particular classes. Making claims about the universal applicability of this model assumes a certain homogeneity of behaviour among all speakers and we are forced to represent the Model Speaker as ‘the embodiment of sharedness’, that is speakers and hearers are represented as sharing values and beliefs about what is appropriate language (Eelen, 2001: 132). The way that individuals negotiate with their perceptions of politeness norms is thus lost. For example, one of the issues which came up frequently in the interviews I conducted was that many of the interviewees recognised different levels of politeness as being habitual, constituting part of the style for particular people. It is clear that many people recognise that some people have certain linguistic stances which mean that their behaviour is consistently over-polite or impolite compared to the community of practice within which they are participating, or even perhaps, in general, in all of the situations in which they are involved. For example, in discussing politeness with a group of lecturers in Sweden, one female lecturer stated that one of her male colleagues, who was present at the seminar, always talked to her in a very brusque way, and she stated that she did not like having to work out what he meant rather than taking this brusque style at face value. In other words, she resented him for not providing the level of politeness that she expected of him, and she felt that she was having to assume positive politeness was

Theorising politeness

107

intended by her colleague even though it was not expressed; thus, all of the interactional work and reasoning was being left to her. The male colleague said in response: ‘Well, you know me, you should know what I mean.’ Thus, the male colleague assumed that, because of the level of familiarity between them and because he consistently used this style, she should not take offence. To give an example from my data, in example 4, C, an older, white, upper middle-class, ‘out’ homosexual male is consistently impolite, when judged against the linguistic behaviour of the others in the community of practice, all of them white, middle-class teachers. However, none of them seems to take offence or react to his rudeness within the interactions they have with him. However, all of them, both before and after this interaction, characterise C as ‘a curmudgeonly old bugger’ or as ‘bloody rude’ and ‘difficult’. This is a judgement about his habitual behaviour, however, and not an assessment of his behaviour in particular interactions (although there is clearly an interaction between these two assessments). Take this example, where C comes into a room where two younger, white, middle-class heterosexual male friends, T and D, are talking: →

1T: 2C: 3C: 4C: 5T: 6C: 7C: 8T: 9D: 10T: 11C: 12T 13C: 14D: 15D:



16C:



17T 18C:

Hello There’s the man who’s obviously got it in for me Christ [(general laughter)] ALMIGHTY (.) if one man could bugger me around more in two days than you have in the last two days I’d like to know his name Christ [(laughs)] Almighty (.) Still all right. If I didn’t know you I would say it was on purpose [ C, have a drink [ Hello C= =Red ros´e or beer? Yes please= =All of them (laughs) Large? [ I’d like another beer= =So what are you talking about (.) the last two days? I haven’t done anything to you in the last two days [ ] That’s precisely what I’m complaining about (laughs) Is your arm all right? No no it’s fine [

108

Gender and Politeness

19D: Why what should I have 20D: done?= 21C: =Strewth well I thought I just thought that you were going to do things 22C: maybe you weren’t (.) and you didn’t Data: 107.1/2 EXAMPLE 4

C, consistently through all of the conversations I have recorded between him and a variety of friends, is far more direct and openly confrontational than they are towards him and he engages in a great deal of verbal play or banter.30 None of the other participants reciprocates with the same degree of fairly confrontational behaviour, although all of them swear in equal degrees to each other (although not at the same level as C). Most of them tend to be more confrontational with him than with other people in other groups or individually with other members of the same group. These people belong to a fairly close-knit group who see each other on a daily basis, both at work and socially. But here D states that he does not know what he is accused of doing or not doing, and even when he asks C what he is being accused of, he is not told, but seems to take this with humour and accepts this as just another example of C’s habitual style. This notion of personal style in relation to politeness is extremely important. Many of these participants regard C as being difficult, using the term ‘prickly’ to describe him, but when they judge whether he is being polite or considerate towards them, it is against the standard of his personal style, rather than against a notion of what is appropriate for most other people in a particular setting.31 These methodological problems with data collection and interpretation have forced me to re-evaluate the way that we analyse politeness. It has made me challenge what we see as constituting data and the claims that we make for these data. It has also forced me to question the way that we draw on the notion of the variable in our analysis, since neither CDA’s analysis of power is sufficient to grasp the complexity of the way that power is negotiated, nor is CA’s insistence that we must analyse only that which is oriented to by the participants in the conversation. In the following chapters I sketch out a form of analysis which I hope is more adequate for the analysis of judgements of politeness made by interactants, which tries to combine elements from CA and CDA, and which tries to

Theorising politeness

109

attend to the concerns of the participants whilst also being aware that those concerns are not entirely of their own making. Implications for analysis This critique of Brown and Levinson’s model of politeness has many implications for analysis which I would like to summarise here. Firstly, it is important to see that politeness has a range of different functions and is not simply an expression of affiliation with, or deference towards, others, even though stereotypically that is the way it is often discussed by both analysts and interactants. Thus, politeness should be seen as a set of strategies or verbal habits which interlocutors set as a norm for themselves or which others judge as the norm for them, as well as being perceived as a socially constructed norm within particular communities of practice. We might see politeness as akin to resources which are viewed differently by interactants and which may be drawn on by them to different extents because of their assessments of their position of local and institutional power relative to others, and because of the way that they themselves are treated by others. Participants make these assessments of how much politeness they need to use on the basis of what they manage to achieve in talk, and sometimes others’ assessments of them and others’ lack of support in conversation may mean that they are forced to use certain polite strategies in order to achieve what they wish, or in order to ingratiate themselves with others. Holmes seems to affirm this in that she talks about ‘polite people’ as those who ‘avoid obvious face-threatening acts . . . they generally attempt to reduce the threat of unavoidable face threatening acts such as requests or warnings by softening them, or expressing them indirectly; and they use polite utterances such as greetings and compliments where possible’ (Holmes, 1995: 5). However, this view of ‘polite people’ does not relate those polite acts to a community of practice which judges the acts and the people as polite, or which sees that polite behaviour as determined by the actions of people other than the individuals themselves. Thus, this is again an example of the disembodied, abstract, analysis which is often determined by the use of Brown and Levinson’s framework. Politeness can only be analysed within particular communities of practice and should be seen as negotiations with assumed norms. I am not assuming that

110

Gender and Politeness

the norms of communities of practice are uncontested, for as I have tried to show in the analyses of data in this chapter, there are generally misunderstandings and disagreements about what is going on in conversations and what is appropriate. As I have stressed throughout, the notion of what is appropriate is open to debate within communities of practice and is often something which is contested: ‘The phrase appropriate or nonproblematic advertises its own roots in collectivist homogeneity, in which each knows the (single, determinate code-like) language perfectly. In practice, all these homogenised singularities strive to conceal pluralities. There is almost never a single appropriate or nonproblematic use of any particular word but always a great number of ways of using it appropriately or inappropriately. And on each distinct occasion of use, it will be some of these endlessly varied judgements of appropriateness that will be displayed by interlocutors in situ’ (Toolan, 1996: 170). The notion of hypothesised appropriateness in the assessment of an act as polite refers to the judging of whether an utterance is appropriate or not, either in relation to the perceived norms of the situation, the community of practice, or the perceived norms of the society as a whole. There is obviously flexibility in these norms and the potential for misunderstandings and misapprehension of politeness is great. For example, in an anecdote given by one of the interviewees, a male public speaker leading a workshop for a group of male and female professionals used mild swear words and a range of informal expressions in order, one assumes, to set the group at ease and create an atmosphere of informality and openness (that is, he intended to pay positive politeness to the face needs of the group, and assert that he wished to be viewed as a colleague rather than as an expert), but this was interpreted by some of the group members as impolite, ingratiating, or patronising, since they had clear views of the language which they felt appropriate to the professional setting and the role of workshop discussion leaders. They felt that not enough respect was being paid to the community of practice as a whole and to them as individual participants. The second implication of these difficulties with Brown and Levinson’s model is that, in analysing politeness we must constantly remind ourselves that politeness is a matter of judgement and assessment, rather than politeness residing in particular linguistic forms or functions. And thus, the hearer has to assess at all times the

Theorising politeness

111

commitment that the speaker has to his or her statement and whether they are sincere or merely being manipulative. Speakers and hearers are constantly assessing the interaction in relation to politeness norms that they assume are operating in the community of practice. Taylor argues that this judgement has a moral dimension: ‘language appears not as an autonomous system of formal regularities, but as a normative practice, the regularity of which we ourselves create, police and reward as a part of the very performance of that practice, and to which we attribute what amounts to a moral value’ (Taylor, 1992: 13). Toolan argues that, in contrast to the centrality of practical reasoning in the model of language as a whole and in politeness analysis in particular, we should instead focus on the role of imagination, since it is precisely this creative empathetic skill which is drawn on to make sense of others: ‘imaginative powers enable us to make rational assessments of what – given our awareness of the present circumstances and present agendas of interactants, together with awareness of how perceptibly related gestures, signs and effects have been interpreted on past occasions (i.e. memory) – might possibly be going on in a current situation’ (Toolan, 1996: 177). As I discussed earlier in this chapter, there are no formal constitutive elements of, for example, an apology in isolation; Toolan argues that: in practice in heterogeneous speech communities, the relations between specifiable conditions and specific acts invariable turn out to be many to many: rather different conditions can count as enabling what particular community members would regard as an apology, and rather different acts can count as enabled, for particular members, by any particular set of conditions. . . . What is constitutive of an apology, or a threat, or a warning, is just that complex of situated and cotemporal behaviour . . . experiencing that an addressee concludes that he or she has been apologised to, or threatened or warned. (Toolan, 1996: 299)

Whilst the phrases ‘I am sorry’ and ‘I apologise’ are often considered by linguists to function as apologies, if the hearer considers that the speaker is insincere when s/he utters them, they simply will not function as an apology at all. Apologies are often composed of elements which cannot be recognised easily by either interactants or analysts as unequivocal apologies; for example, following on from the interaction, in example 4, discussed earlier between C and a group of friends, where C accuses one of them, D, of not doing something for

112

Gender and Politeness

him, D then, thirty-five minutes later in the conversation asks for clarification of his earlier accusation, after they have been talking about a range of work-related issues: all I want to know is why you’re so fucking ANGRY with me= it’s not really a question of being angry but I’ve been feeling particularly sick it might have been that [ 4D: well what has that got to do with me? = 5C: =well I was feeling sick it must have been this magical communion you were having I’ve only just got down to its reason= 6D: [(laughs)] 7D: =I wish you well (Data 107.5) EXAMPLE 5 →

1D: 2C= 3C:

Here, long after the initial accusation, D seems to be trying to resolve the matter, or at least trying to force C to explain his attack on D. C seems to be trying to suggest that the problem did not lie with D at all, but was rather due to his feeling unwell, →

2C= 3C:

it’s not really a question of being angry but I’ve been feeling particularly sick it might have been that

and D’s laugh and comment in line 7: ‘I wish you well’ seems at face value to be recognising this as an apology for C’s earlier outburst (although it has to be said that D’s comment cannot be heard as unequivocal, since it could be interpreted ironically or sarcastically). Spencer-Oatey argues that ‘rapport threat and rapport enhancement are subjective evaluations, which depend not simply on the content of the message, but on people’s interpretations and reactions to who says what under what circumstances’ (Spencer-Oatey, 2000: 19). Thus, an apology is not necessarily simply a formal linguistic unit, but is a judgement made about someone’s linguistic performance: whether the right amount of effort and work has been expended; and whether sufficient commitment and sincerity have been expressed. I have argued that the question of judgement should be central to all analyses of politeness. We can see this process in action in the following extract, where two colleagues in a university are analysing what constitutes politeness or impoliteness. Here, two heterosexual, middle-class, white male lecturers, J and G, are discussing a situation which occurred in a university where they were both teaching; the discussion seems to centre on the question of how to evaluate the behaviour of others:

Theorising politeness

113

1J: 2J: 3J:

It’s not even politeness(.)cause I was saying to my class. my class wanted to walk out(.) and I met A so I wanted to talk to him about the course business you know he IS head of department and I hadn’t seen [ ] 4G: Mmmhmm 5J: him for a long time(.)and I talked for perhaps too long(3)then his role 6J: was taken over by (.) the head of philosophy(.)who seemed to be you 7J: know this old power business Then I saw my class just drifting out 8G: [Mm] 9J: I’d only been out there for x minutes(.) I said where are you going 10J: (.)they said well we are fed up with waiting for you having a good old 11J: natter(.)which I’ve never done before(.)I said what do you want to do 12J: create problems or solve them(.)you’re bothered that I’ve talked for so → 13J: long outside the door(.)I said the reasonable way to solve them is to 14J: say excuse me I said I was obviously wrong(.) I shouldn’t have done that 15G: [Mmm] 16J: it was not intentional.it could have been solved by talking not a great 17J: confrontation protestation= 18G: =Yeh= 19J: =and then when I talked to them like that they all came back(.)you 20J: know I said that’s great you know (2)I’d like to talk to you you talk to 21J: me and the problem’s solved and then once in the whole time since I’ve 22J: been here I talked too long outside the door. and instead of talking to 23J: me(.) no all walk out But I suppose from young people But I suppose 24G: [Mmm] 25J: from young people that’s the way that they handle things= 26G: =it’s entirely possible that they would see it not(.) er as a 27G: confrontational thing um(.) they probably feel that if they came up to 28G: you and said look er hum you know [ 29J: Yeh I’m (.) I’d have given a smart aleck reply [ 30G: Well(.)or 31G: sort of. tear them off a strip(.) or whereas just= 32J: =yeh= 33G: =sort of = 34J: =yeh yeh I can yeh I thought of that Data: 104.7/6 EXAMPLE 6

Here J seems to be asking G to try to help him to work out a problem with the behaviour of a class whom J perceived to have behaved impolitely towards him and a great deal of the talk is expended on assessing the way that they had behaved impolitely and what they should have done in his opinion in order to be behaving towards him appropriately (lines 13–25). G, to resolve this problem, seems to suggest that their behaviour could have been motivated by other factors which overrode the imperative of deferential politeness towards their lecturer, for example that they were afraid of being mocked, which J suggests might be the case:

114

Gender and Politeness 26G: 27G: 28G:



29J: 30G: 31G: 32J: 33G: 34J:

=it’s entirely possible that they would see it not(.) er as a confrontational thing um(.) they probably feel that if they came up to you and said look er hum you know [ Yeh I’m (.) I’d have given a smart aleck reply [ Well(.)or sort of. tear them off a strip(.) or whereas just= =yeh= =sort of= =yeh yeh I can yeh I thought of that

G does not explicitly offer an explanation for this, choosing rather to refer to an unstated but assumed to be mutually manifest knowledge of J’s past behaviour by saying ‘you know’ (line 28). (This ‘you know’ may be the result of diffidence on G’s part about seeming to accuse J of losing his temper with students, or it may simply be indicative of a need on G’s part to pause and think what the students’ behaviour means.) Thus, as we can see from this interaction, judgements about what behaviour constitutes politeness or impoliteness are often achieved with others rather than through a simple process of individual judgement. This is necessarily a moral process: ‘the individual’s active and reflective part in language use is characterised by speakers’ unequivocal habitual sense that they are personally and severally responsible for what they say and for the effects that their uses of language have. Morality is involved since in practice language is always in use within human purposeful activities, and those are inescapably value-laden, choice implicating, preference and assumption reflecting, reflecting and articulating judgements of what is right and wrong, desirable and undesirable, correct and incorrect. Part of the standard depiction of the language user has entailed a theoretically absolute separation of this value-ladenness, this concern for correctness, that is implicit and explicit in everyday language use’ (Toolan, 1996: 179). Since this is a process of judgement, there can be misunderstandings or disagreements about what constitutes politeness or impoliteness. For some people, a particular FTA might be offensive and constitute a breach in a relationship; for others it might be simply part of the give-and-take of relationships. In analysing data, this must be reflected, and in the analyses that I have undertaken in this chapter I hope I have shown that there is a range of possible interpretations. This move to the analysis of judgement necessitates

Theorising politeness

115

a focus more on the discourse level of analysis than on individual words. Rather than reifying politeness and assuming that politeness is a material entity, we need to focus on the way that perception of what constitutes politeness or impoliteness structures interaction and is drawn on explicitly by interactants, and functions to structure their roles in relation to one another. The third implication of these difficulties with Brown and Levinson’s model is that because of the difficulties which I discussed in chapter 1 in relation to gathering, analysing, and interpreting data, and also because of the difficulties which I discuss in chapters 3 and 4 in relation to generalisations in relation to gender, I would argue that different forms of data need to be considered. As I mentioned in the Introduction, I have decided to use both conventional forms of transcribed data, together with anecdotes which I have noted down or which have been elicited from interviews with a wide range of people. Although I note the difficulty with early feminist linguists’ over-reliance on anecdote, as Cameron has remarked, there are times in theoretical discussion when an anecdote can strategically point up certain types of linguistic assumptions: ‘stereotypes . . . condense a great deal of taken-for-granted cultural wisdom into a very small amount of surface discourse production. It is precisely the relationship of the surface utterance to the cultural assumptions “underneath” that is at issue’ (Cameron, 1998a: 447). Beebe argues that there is a place for using a variety of data, examples which have been noted down immediately after the incident took place and reconstructed examples which have been written down some time after they took place. She argues that this type of ‘notebook data’, although problematic in that the type of data is restricted to the circle of friends, acquaintances, and the strangers that one meets, has the advantage of being spontaneous (Beebe, 1995: 158). She argues ‘the gain from removing limits on subjects and settings is tremendous in a study which is aimed at taxonomy – especially classification of a behaviour [such as impoliteness] which is stigmatised and therefore likely to be edited or distorted if elicited and likely to be absent in many formal settings conducive to traditional scientific data collection’ (Beebe, 1995: 158). One could also argue that since the analyst was involved or knows the people involved there is a level of knowledge about the interactions which could not be achieved with data from respondents in questionnaires or from laboratory conditions.

116

Gender and Politeness

Because politeness is a process of judgement, it is precisely this process which must be analysed and anecdotes about incidents which were judged to be impolite or polite serve this purpose. Conclusions In criticising Brown and Levinson’s model of politeness and their methodology, I have been trying to formulate a form of analysis which is adequate for the analysis of politeness. What we should aim for is a form of analysis which tries to uncover the motivations and interests of interactants in particular contexts. ‘The integrationist seeks above all an “inward” account of language, as opposed to a detached, abstracted and idealised one’ (Toolan, 1996: 22). What Toolan argues we should be trying to find out is whether ‘what is modelled is truly the lay language user’s own understanding of the given phenomena’ (Toolan, 1996: 22). Thus, Brown and Levinson’s model, because of its idealised nature, cannot deal with the way that politeness operates in real conversations as a form of assessment of behaviour. In discussing Brown and Levinson’s work, Kasper suggests that their work and that of other theorists is ‘impressive in their parsimony and elegance [but their models] are over-simplistic. Their lasting achievement is to have produced excellent heuristics to investigate a complex object of inquiry. As theories with claims to universality, they need elaboration and revision’ (Kasper, cited in Keinpointner, 1997: 254). However, I feel that elaboration and revision cannot solve the problems in Brown and Levinson’s work. Perhaps what I am arguing for is, as Held puts it, ‘an evolutionary interactive concept [of politeness] which is dependent on reactions by and/or the understanding of the co-interactant’ (Held, 1992: 146). It is this interactiveness, particularly in relation to the community of practice and the context, that I see as playing a major determining role in the production and assessment of politeness, and that requires different forms of analysis. Notes 1. I would like to thank Chris Christie and Clare Walsh for their invaluable comments on this chapter. 2. Brown and Levinson themselves, in the introduction to the second edition of their work, are very self-critical, refining the type of theoretical and analytical work that they suggest that they need to draw on. However,

Theorising politeness

3.

4.

5.

6.

7.

8.

9.

10.

11.

117

in essence, their analysis of politeness remains much the same as in the 1978 publication, even though they consider certain theoretical revisions advisable. If they seriously questioned their own use of Speech Act theory, this would necessitate a complete revision of their analytical model. However, as Diamond has shown, this notion of a sequence of Face Threatening Acts (FTAs) followed by repair does not happen in practice, since the supposed repair very often precedes the FTA or is interwoven with the FTA (Diamond, 1996). I would disagree with this assessment that it does not matter at all whether the apology is perceived to be insincere, but I would agree that it is perhaps more important that an apology is publicly and clearly made. Etiquette has an interesting history, which may explain this sense that etiquette is outdated and should not be analysed in conjunction with politeness. Ehlich argues that politeness, in the guise of formal etiquette, only played an important role in the French court among the aristocracy because they were no longer in such a powerful position; hence the sense now that etiquette is an outmoded and socially useless form of politeness (Ehlich, 1992). As I mentioned in the Introduction, I conducted taped interviews with a wide range of people, based on a questionnaire which tried to elicit judgements about politeness in general, and anecdotes about conversations where politeness or impoliteness had seemed to the interviewee to be problematic or excessive. It might be argued that this holding of these opposing views on politeness demonstrates the complexity of their position; however, I would suggest that these views of politeness are only implicitly articulated and thus can only be seen as contradictory. However, this analogy forces us to assume that, like money, politeness has a value which is agreed on by all participants. There is generally a distinct lack of agreement about whether a particular utterance can be considered to be polite or not, and whether polite behaviour is trivial or valuable. For example, we might consider the fact that students do not tend to talk in lectures and the assumption that if they do, they are threatening the face of the lecturer. There is no regulation that students should not talk in lectures, and individual students do not necessarily go through a process of deciding not to talk; it is an assumption which is constructed at the level of the community of practice that it is in everyone’s interests if students do not talk. It is debatable whether Brown and Levinson’s model, with its emphasis on the Model Speaker and his/her strategic language use, could in fact integrate any notion of social politeness. However, as Eelen (2001) remarks, this metaphor explains why people abide by the rules, but cannot explain what motivates people to break them.

118

Gender and Politeness

12. However whilst Coulmas’ analysis of Japanese is accurate, it is also the case that other commentators have noted that certain groups of Japanese speakers have managed to challenge the use of these seemingly essential honorific particles as part of a process of revising their assessment of their position in the social order (Okamoto, 1995). 13. Some of them remarked on particular children’s use of swear words which they considered impolite, and which the children continued to use despite interventions by teachers. 14. However, Watts does not seem to be able to account for the fact that the use of ‘sir’ in situations where it would not normally be expected might be interpreted as ironic or offensive. 15. The difference between Janney and Arndt on the one hand and Watts on the other is less to do with behaviour and more to do with their different perceptions of what motivates people to be polite in given situations (Walsh, pers. comm.). Although I find Watts’ distinction between what is polite and politic interesting in drawing attention to the possibility of using polite forms in manipulative and egocentric ways, because this distinction is phrased in a slightly counter-intuitive way, I shall not be maintaining his distinction in my analysis. However, I retain the notion that individuals may use over-politeness as a way of trying to accrue benefit to themselves. They need not be successful in this, however, as others may see them as manipulative or as having assessed social distance incorrectly. 16. Very few theorists consider the historical development of politeness, but most of them, because they consider the origin of language to be from a very basic and ‘primitive’ form of speech associated with grouporganised activities, would probably assume that politeness only developed fairly late in human evolution with the development of ‘civilisation’ (Beaken, 1996; Foley, 1997). 17. However, Jary argues that there is no single ‘message’ of polite behaviour: ‘much of what has been termed polite behaviour . . . is best seen as directed at avoiding unwanted implications rather than communicating implications. Moreover, when speakers do intend their behaviour to have certain implications, they often want these intentions to remain hidden from their hearers as their recognition would result in the speaker’s objectives not being achieved’ (Jary, 1998: 13). Thus, politeness, in this view, is more like a clarificatory form of behaviour, or a form of repair work, than a substantive form of linguistic communication in its own right, and Jary illustrates the complexity of assessing what exactly someone is doing when they are using language which might be assessed by others as polite. 18. Just because someone has assessed the level of appropriate behaviour does not necessarily mean that they will behave according to these hypothesised norms. However, their notion of appropriacy will play a role in assessing their and others’ behaviour as aberrant, impolite, and so on.

Theorising politeness

119

19. This may also be the case with Australia and America, where, at a stereotypical level, blunt, direct, speaking is preferred to the class-based British norms of politeness, and this type of direct speech may have been consciously adopted as a way of distinguishing these cultures from Britain. 20. The Cross-cultural Linguistic Politeness Research Group, composed of linguists from Britain, China, Georgia, Libya, Italy, Turkey, Finland, Egypt, and the Netherlands, has been collaborating on rethinking the models which are currently in use for the analysis of linguistic politeness. We meet regularly to discuss the research of the participants and also to discuss new research in this area. One of the main discussions so far has been on the contestation of the notion of face and communities of practice and politeness. Details of the group can be found on the website at http: //politeness.lboro.ac.uk or by contacting Sara Mills: [email protected] or Chris Christie: [email protected]. There is a special issue of the electronic journal Working Papers on the Web, vol. 3 (2002) www.shu.ac.uk/wpw on the subject of Politeness and Context which contains papers by members of the research group. 21. And in other analyses of politeness using Brown and Levinson’s work, the sentence level or phrase level analysis is the more common. 22. Brown and Levinson themselves do not propose that linguistic forms themselves are intrinsically polite; it is rather strategies which are polite. However this is equally problematic. 23. The strategic or unintentional use of forms of linguistic behaviour which are interpreted as over-politeness cannot be analysed within the Brown and Levinson framework, because of the focus on the speaker’s intentions alone. 24. However, even swearing, if uttered in a joking voice, may be assessed as positively polite, suggesting equality of relation between boss and secretary, or suggesting a lack of social distance in this particular context – a sense that both people are united against a common problem. 25. A strategy frequently used by those who are considered by others to be rude or brusque. 26. This distinction between an analyst imposing a meaning on an utterance and an analyst attempting to discover the meanings which interactants give to an utterance is one which Bucholtz (1999a) defines as the distinction between sociolinguistics and ethnography. And Schegloff (1997) would see this as a crucial distinction between CDA and CA. However, many ethnographers still confuse the position of the interactants with their own analytical position, as is shown in Schegloff’s analysis of conversations where he assumes that he has captured what was really going on in an interaction (Eelen, 2001). 27. Brown and Levinson also note the difficulty of assigning certain utterances unequivocally to a category and fitting every utterance within their framework. They argue that, with a strategy like ‘Be pessimistic about the success of an FTA’: ‘an utterance like “You don’t want to

120

28.

29.

30. 31.

Gender and Politeness pass the salt” should be polite: that it is not, of course is due to the fact that it attributes impolite desires to the addressee’ (Brown and Levinson, 1987: 11). They argue that this ‘over-generation’ of utterances as polite could be compensated for by building into the system a set of ‘filters’ which would check that there were no impolite implicatures. However, this rather defeats the object of developing a general framework. This is not an isolated example of an utterance failing to fit the model, but rather a general problem with the model itself where complete comprehension and clarity is assumed. Scott (1990) has analysed the silence of those who are in a position of inferior status in conversation describing their silence as the ‘official transcript’ and has contrasted this with what he calls ‘the hidden transcript’, that is a critique of those power relations articulated among equals when the powerful person is out of sight. He asks therefore ‘How do we study power relations when the powerless are often obliged to adopt a strategic pose in the presence of the powerful . . .?’ (Scott, 1990: xii). From this, it is clear that silence cannot simply be equated with powerlessness since it is an adopted pose which needs to be analysed alongside the ‘hidden transcript’. This view was also articulated in a BBC Radio programme (2001) on perceptions of the importance of politeness by older and younger people in Wales. I discuss banter in more detail in chs. 3 and 4. Ultimately, I cannot tell exactly what was going on in this interaction at an interpersonal level since I was not present; thus, the smiles, avoidance of eye contact, and physical position of interactants cannot be analysed. Even though I talked to the participants about the interaction afterwards, and know all of them very well, it is impossible for me to know what exactly is being negotiated interpersonally here. The exact nature of the continual rivalry between C and D, which occurs on other occasions, in particular is not clear to me.

3

Politeness and impoliteness

Introduction There has been surprisingly little analysis of impoliteness itself, in research on politeness in general; perhaps this can be attributed to the fact that much of the research is dependent on a view of conversation which ‘emphasises the harmonious aspect of social relations, because of an emphasis on conversational contracts and the implicit establishment of balance between interlocutors’ (SpencerOatey, 2000: 3). However, there are occasions when people attack rather than support their interlocutors, and sometimes those attacks are considered by others to be impolite and sometimes they are not. Keinpointner argues that non-co-operative behaviour should be seen as less exceptional than most politeness theorists see it (Keinpointner, 1997). He suggests that it is idealistic to assume that everyone tries to co-operate for most of the time.1 However, Eelen argues that the model of politeness drawn on by researchers in this field is one which implicitly or explicitly focuses only on politeness and sees impoliteness as a deviation; this causes theoretical difficulties since ‘the concepts involved can never explain impoliteness in the same way or to the same extent as they explain politeness. So the polite bias is not just a matter of differential attention, it goes far deeper than that: it is a conceptual, theoretical structural matter. It is not so much quantitative, but rather a qualitative problem’ (Eelen, 2001: 104). Furthermore, the polarisation of politeness and impoliteness might lead us to assume that, for interlocutors, behaviour falls into either one or the other category. Although this is correct up to a point, speakers and hearers may be generally tolerant in relation to making judgements about whether an exchange is polite 121

122

Gender and Politeness

or impolite, accepting statements which may be a little ambiguous in terms of their function, as part of the give-and-take of interaction. It seems to be only at moments of crisis that judgements about impoliteness are made. This chapter investigates the ways in which politeness and impoliteness have been described, focusing on clear differences between them. Rather than assuming that there is something intrinsically impolite about certain utterances or exchanges, I argue that impoliteness is attributed to a speaker on the basis of assessments of their intentions and motivations. I examine impoliteness in its own terms, rather than in terms of its relation to politeness, considering what factors contribute to the assessment of an act as impolite, and what consequences the judgement of impoliteness has on individuals and communities of practice. Thus, firstly I question the notion that politeness and impoliteness are binary opposites. I analyse the factors which lead to judgements of impoliteness, and then consider, in particular, the role hypothesised stereotypes of class, gender, and race play in assessing impoliteness. Finally, I analyse an incident which I considered to be impolite; the analysis of this incident exemplifies the process of judgement of appropriacy on the basis of hypothesised stereotyping within a particular community of practice. Politeness and impoliteness In contrast to a great deal of research in this area, I believe that impoliteness has to be seen as an assessment of someone’s behaviour rather than a quality intrinsic to an utterance. Many theorists, following Brown and Levinson, assume that impoliteness is necessarily an attack on the ‘face’ of the interlocutor/s, and that ‘certain “impolite” speech acts, such as reproaching, threatening and insulting are performed by speakers with the intrinsic purpose of attacking or undermining the hearer’s face’ (Haverkate, 1988: 394). The analysis of impoliteness is therefore concerned with a reconstruction of what the speaker’s intentions are supposed to have been. Culpeper questions Leech’s notion that there are some speech acts which are inherently impolite, and suggests that although there may be a few, they are in the minority (Culpeper, 1996). Only those acts which do not orient to a virtual or potential offence, and which are offensive in themselves, should be seen as inherently impolite, he argues; all other

Politeness and impoliteness

123

acts should be seen as contextually relatively impolite. However, we might question that any act is necessarily intrinsically impolite, since even the most offensive insults can be used by close friends to signal camaraderie. Lycan questions the notion that speech acts such as interruptions, even when they are bald-on-record interruptions, are necessarily interpreted as face-threatening, and he draws attention to the fact that in certain types of academic discussions, for example among philosophers and linguists, interruptions are, in fact, seen as positive contributions to the development of the discussion (Lycan, 1977: 24).2 Lycan suggests it is simply ‘prudish’ – an interestingly gendered term in this context – to assume that interruptions are, in essence, impolite. It may be the case that certain acts are associated with impoliteness; for example, with speech acts such as threats, in certain contexts where it is clear to both speaker and hearer that the speaker intends to threaten the other, but this is rarely the case, since most of the time there is an option of understanding the utterance in another way (considering it as a case of misunderstanding on the part of the hearer, for example, because of over-emphasising the importance of certain cues), or of considering that the threat is in fact better interpreted as a case of accidental, or unintended impoliteness (that is, a fault of expression on the part of the hearer).3 Keinpointner (1997) distinguishes between motivated and unmotivated rudeness; in motivated impoliteness, the speaker is assumed to have intended to be rude, whereas unmotivated impoliteness is the result of insufficient knowledge of some kind. Thus, hypothesising of intention is essential to assessing an act as impolite. Keinpointner (1997) analyses the variety of acts which can be classified as impolite, ranging from mock impoliteness or banter, where intimates insult one another, but do not take offence; to strategic rudeness, where the insults are non-reciprocal, often within an institutional context, and which may or may not cause offence. However, many of the categories of rudeness that he suggests are only impolite in some contexts, or not clearly impolite. For example, as Culpeper points out, mock impoliteness only functions to foster social intimacy when it is clear to all parties that the impoliteness is untrue (Culpeper, 1996). However, perhaps it could be considered that mock impoliteness might actually be used, precisely because there is an element of truth in the utterance. Yedes argues that what she terms ‘playful teasing’, in a work environment, allows

124

Gender and Politeness

conflicts to be resolved and tension to be managed (Yedes, 1996). She shows how, in order to get someone to volunteer for unpleasant tasks, insults and jokes are used to manage difficult situations where refusal to find someone to do the tasks would cause disruption to the community of practice as a whole. Yedes argues that ‘teasing mitigates conflicts, reaffirms affiliation and encourages equity in relationships’ (Yedes, 1996: 418).4 Thus, banter or mock impoliteness might allow someone to utter something closer to their true feelings in an exaggerated form at the same time as posing it in a manner where it will be interpreted on the surface at least as non-serious. Thus, impoliteness is a very complex assessment of intentions and not a simple either–or category, as some theorists seem to suggest. Keinpointner discusses Schiffrin’s (1984) research on the interactional styles of Jewish couples in Philadelphia, where ‘strategies of communication which would be experienced as aggressive, non-co-operative behaviour by other groups of the Anglo-Saxon speech community, [such as] preference for disagreement, increased volume, rapid tempo, persistent attempts to get the floor [are viewed by the interviewees] as a means to enhance sociability’ (Keinpointner, 1997: 268). However, we might question that this is, in fact, impolite behaviour, if the interviewees themselves do not regard it as such. It would only be considered as impolite if viewed by someone from another language group whose politeness norms were significantly different. The evaluative nature of assessments of impoliteness is particularly striking in experimental situations. Eelen describes the difficulties which arise when interviewees are asked about the type of response that they would give in particular situations in role plays or gap-fill exercises, as politeness and impoliteness are often considered to be concerned with proper and improper behaviour. He argues ‘informants show a clear understanding of “right ” and “wrong” behaviour, and they always situate themselves, not really surprisingly, on the “right” side’ (Eelen, 2001: 39). He goes on to suggest that impoliteness is therefore assumed to be a quality associated with other people’s behaviour and never one’s own. Politeness and impoliteness cannot therefore be considered to be simply polar opposites. Culpeper (1996) takes Brown and Levinson’s four super-strategies (bald-on-record, positive politeness, negative politeness, and off-record) and inverts them to describe

Politeness and impoliteness

125

impoliteness: thus, he analyses impoliteness as consisting of bald-onrecord impoliteness, positive and negative impoliteness, and sarcasm or mock politeness (Culpeper, 1996). However, Beebe (1995) has shown that this assumption that impoliteness is the opposite of politeness cannot hold; she gives examples of what she terms ‘pushy politeness’, where seemingly polite utterances are taken to be impolite and face threatening. She analyses an incident in a busy New York restaurant, where a group of people were repeatedly asked by different waiters if they would care to order, when they had, in fact, made it plain that they wished to have a discussion over lunch and would therefore take their time over their meal. The attentiveness of the waiters ‘seemed to reflect a desire on the part of the waiters to get it over with, not a policy regarding length of stay, a lack of communication among waiters, or a problem with crowding, so it was viewed as rude’ (Beebe, 1995: 161). Keinpointner also draws attention to the fact that some forms of politeness, such as manipulative or insincere politeness, should be seen as less than optimally co-operative or rational, and hence impolite (Keinpointner, 1997). Thus, politeness and impoliteness cannot be seen as simple polar opposites. Nor should impoliteness be seen as the marked term, in relation to an unmarked norm of politeness, since this assumes that politeness is almost invisible because it is the norm, whilst impoliteness is noticeable. Furthermore, the description of impoliteness should not simply be phrased in evaluative terms where impoliteness is treated as the ‘abnormal and irrational counterpart of politeness’ (Keinpointner, 1997). Thus, rather than a simple opposition between politeness and impoliteness, Keinpointner suggests that we should try to consider linguistic behaviour along a continuum, as a matter of degree rather than absolutes. Whilst agreeing with this notion of a continuum, it is important that we see this as a continuum of assessment rather than a quality of impoliteness and politeness. An important aspect of the evaluation of utterances as polite or impolite is the degree to which institutions have routinised the use of certain types of language. Thornborrow argues that institutions tend to constrain what can be counted as a legitimate contribution and also the ‘discursive resources and identities available to participants to accomplish specific actions are either weakened or strengthened in relation to their current institutional identities’ (Thornborrow, 2002: 4). In his analysis of impoliteness, Culpeper (1996) analyses

126

Gender and Politeness

several contexts of linguistic usage – a documentary programme on American army training and literary drama – where he isolates certain examples which he suggests are impolite linguistic behaviour. In his analysis of the army training documentary, he lists several instances of impoliteness by the trainers to the recruits: the trainers swear at the recruits and humiliate them by calling their competence into question. Direct commands are given without any mitigation and formulaic politeness, such as the use of ‘please’ and ‘thank you’, does not feature at all. Culpeper argues that the reason such ritualised insults are used is to train the recruits into accepting their place unthinkingly in the army hierarchy, so that they will obey orders. Politeness is used as a resource to indicate that one acknowledges the interlocutor as a human being, as part of a shared community of practice, whereas the army attempts to deny their basic humanity through the use of language which would normally be considered extremely impolite: ‘in the context of the army, impoliteness is not a haphazard product of say a heated argument, but is deployed by the sergeants in a systematic way as part of what they perceive to be their job’ (Culpeper, 1996: 359). To give an example of the level of ‘impoliteness’ used in this context, one of recruits, Private Alves, a woman of Mexican origin, is told ‘you don’t even deserve to live in the United States’ and she is told that she is a ‘disgrace to the uniform’; the sergeants suggest that she is mentally ill and one says to her: ‘do me a favour don’t have any children . . . because unfortunately there is such a thing as hereditary genes that I would hate to think that anybody would even closely come out like you’ (cited in Culpeper, 1996: 359). These racist insults would in any other context be seen as face threatening. However, I would argue that within this particular community of practice, this behaviour is not necessarily classified by any of the participants as impolite. The dominant group in the interaction, the officers, as representatives of the army, are drawing on ritualised and institutionalised codes of linguistic behaviour which have made excessive impoliteness on the part of trainers the norm. It might be the case that recruits may consider the level of verbal abuse over-aggressive and therefore might lodge an official complaint about it.5 But in general, recruits recognise that these forms of speech are simply part of the discourse genre of that particular community of practice. Thus, if we simply analyse impoliteness in the decontextualised way that Culpeper does, we shall

Politeness and impoliteness

127

be unable to grasp the way that impoliteness is only that which is defined as such by individuals in interaction with the hypothesised norms of the community of practice; even here, it is something which may be contested by community members, either openly in the case of complaints, or tacitly, by people who resent the behaviour but do not complain. Another institutional context where insults are used in a ritualised manner is the House of Commons, especially the language used in Prime Minister’s Question Time. Harris (2001) asserts that although much of the discourse of Prime Minister’s Question Time is composed of intentional and explicitly face threatening (or face-enhancing) acts and [. . .] these can be analysed in terms of both the propositional (e.g. hostile/ supportive proposition/presuppositions which preface or are built into questions and responses to questions) and the interactional (e.g. modes of address, turn-taking “rules”, non-verbal and paralinguistic behaviour) levels, in fact utterances which attack the face of the Prime Minister are often evaluated in terms of the efficacy of their attack, rather than their intention to wound and threaten her/his face. (Harris, 2001: 15)

Shaw describes the way that an adversarial style has become normalised within this particular community of practice and failure to adhere to the ‘rules’ of confrontational and aggressive Parliamentary debate may lead to MPs who use different speech styles being admonished explicitly by the Speaker (Shaw, 2002).6 Thus, verbal and argumentative skill is valued within this context in a similar way to the evaluation of ritualised insults in groups of young black males described by Labov in the 1970s, or verbal dexterity and excessive, sexual lyrics in rap music (Labov, 1972; see for a discussion, Matsuda et al., 1993). Harris remarks that jeering and cat-calling in the House of Commons frequently interrupts speakers and ‘this type of interactive behaviour would be unacceptable in most other formal institutional contexts – courts, classrooms, conferences, televised debates – especially those which have explicit turn-taking rules, and either sanctions would be invoked or a marked breakdown in communication would occur’ (Harris, 2001: 25). In this community of practice – those attending the House of Commons’ Question Time – clearly intended threats to face have become institutionalised and are seen as acceptable, partly because of the ritualistic nature of the debate, and the honorifics (such as ‘the right honourable member’, and so on) which accompany this form of abuse. Insults falling outside

128

Gender and Politeness

the clearly demarcated boundaries of what is deemed appropriate are highly stigmatised, leading to ritualised apologies and possibly exclusion of the speaker from the House of Commons. Harris concludes by saying that ‘systematic impoliteness is not only sanctioned in Prime Minister’s Question Time but rewarded in accordance with the expectations of Members of the House (and the overhearing audience) by an adversarial and confrontational political process. Hence, even the most serious face-threatening acts rarely, if ever, occasion a breakdown in interpersonal relationships nor are they intended to’ (Harris, 2001: 32). In this sense, perhaps we could argue that this form of linguistic behaviour, even though it would be impolite in any other context, is not considered to be impolite in the House of Commons. Thus, I would suggest that in certain institutionalised contexts where highly masculinised forms of linguistic behaviour have become ritualised, dominant community members may use insults without being considered to be impolite, since their use of this type of language may be seen as an instantiation of their role within the hierarchy. Within certain institutional contexts such as schools, the domineering linguistic behaviour of teachers may not be considered impolite by pupils, although they might not like the way they are talked to. For example, in this analysis of a discussion between a white, male, middle-class headteacher [E] and two white, working-class boys of eight and nine [L and A] who have been caught fighting, it is difficult to assert that the headteacher is being impolite in his use of direct forms, nor is it easy to say that the boys are being impolite in their short uncommunicative answers. It is more the case that they seem to recognise the type of routines which are required here, but they may differ in the way that they evaluate their own interventions: →



1E: 2L: 3E: 4E: 5E: 6L: 7E: 8A: 9E: 10E: 11E:

you tell me your story well I accidentally kicked him in the face= =how can you accidentally kick someone in the face? yesterday it was that you accidentally punched someone in the face(.) how did it happen? [someone pushed me and I fell over and hit him in the face A, turn round and look at me. What do you say happened?= =he come up to me and just kicked me. [ you need a tissue? have you got a tissue? In that cupboard(.)slide that green cupboard along and you’ll find a tissue. blow your nose. what was you doing to L?

Politeness and impoliteness 12A: 13E: 14E: 15A: 16E: 17A: 18E: 19A: 20E: 21A: 22E: 23E: 24A:

129

nothing nothing whatsoever(.) nothing at all (.) didn’t say anything to him?= → =no= =didn’t do anything to him?= =no he just came along and kicked you? yeh= → =in the face? where was your face then? low to the ground? standing up I was standing up like that how can it have happened as an accident? cause someone push [ 25E: someone pushed you yesterday as well(.) I 26E: can’t believe it two days can I? come on make it easier tell me the truth 27L: I didn’t mean to hit him= 28E: =what did you do then? → 28L: I just I just kicked my foot up and hit him 29L: accidentally= 30E: =you took a kick at him but didn’t mean to actually hit him= 31L: =no 32E: (to A)what do you think? 33A: yeh 34E: does that sound reasonable?= 35A: =yeh= 36E: =you don’t think he really meant to kick you? 37A: no 38E: he was hoping to 39E: miss? [break of ten turns where questioning continues] → 53E: what are you going to ACCIDENTALLY do tomorrow? 54L: nothing= 55E: =sure?= 56L: =yeh → 57E: what are you actually going to ACCIDENTALLY do the day after 58E: after that? 59L: nothing 60E: if it happened tomorrow do you think I’m going to believe → 61E: you? no(.) would you blame me? no(.) If it accidentally happens 62E: tomorrow what do you think I should do? 63L: (2) dunno 64E: well if it accidentally 65E: happens tomorrow that you’re in any fights or any trouble I shall have 66E: to ask your mum and dad to come in and have a word right? Is that a 67E: bruise on the outside as well? 68A: yes 69E: (to L) have you seen the inside of his 70E: mouth? show him(3) what have you got to say? 71L: sorry 72E: do you accept his 73E: apology? 74A: yeh. 75E: all right shake on it then(.) don’t come back tomorrow Data: 61.2 EXAMPLE 7

130

Gender and Politeness

In much the same way as the other communities of practice discussed above, where certain forms of linguistic behaviour are considered to be not so much impolite as an instantiation of the power relations of the participants, here the headteacher and the boys are engaging in a scenario where each of the interactants knows roughly the type of linguistic behaviour that is expected. The institution demands that if children are found fighting then they need to talk to the headteacher and apologise, after a certain ritual humiliation. The head knows that he needs to ask questions (which are mostly rhetorical though not unequivocally so) to establish what happened in order to apportion blame and, finally, to elicit an apology from L and reassurances from both of the boys that they will not fight again. The questions are used in this way as a very powerful resource, as part of the indirect discourses which teachers use rather than using commands or more direct language, in order to deal with managing pupils without bringing about open conflict. Manke (1997) gives the example of a teacher saying to a pupil who is sitting in the wrong seat: ‘Why are you sitting there?’ When the child moves to his proper seat, the teacher says ‘Thank you’, thus establishing that the question was in fact being used as an indirect form of command. The boys know that they need to submit to a certain amount of verbal humiliation before they are allowed to leave the head’s office. Thus, the headmaster uses heavy verbal irony in lines 18–25 to show that the injuries sustained by A could not have been the result of an accident as he claims, and that A also must have played a role in the fighting:





18E: 19A: 20E: 21A: 22E: 23E: 24A: 25E:

he just came along and kicked you? yeh= =in the face? where was your face then? low to the ground? standing up I was standing up like that how can it have happened as an accident? cause someone push [ someone pushed you yesterday as well(.)

We could argue that this verbal irony is lost on A, since E draws attention to the fact that it is impossible for someone to be kicked accidentally in the face when standing up (20E) and A responds by literally detailing the position in which he was standing. However, we could argue that A, by taking the question literally could be understood to be refusing to take part in this apportioning of blame.

Politeness and impoliteness

131

Manke argues that when students refuse to ‘hear’ the indirectness of teachers, they force the teachers to use more direct forms. She states: ‘in this way they force the teacher’s agenda to the surface so they could oppose it and prevent the teacher from maintaining the pretence that what was going on was co-operation and mutual politeness’ (Manke, 1997: 89). Thus, in this case, E is forced to make explicit that he is questioning A’s and L’s assertion that the injuries were sustained accidentally, rather than on the literal level, asking about where A was standing. A similar rhetorical strategy seems to be being used in lines 53–57, where E uses verbal irony to show that L’s claim that the incident was an accident cannot be taken as strictly true: →



53E: 54L: 55E: 56L: 57E: 58E:

what are you going to ACCIDENTALLY do tomorrow? nothing= =sure?= =yeh what are you actually going to ACCIDENTALLY do the day after that?

The boys use only very minimal responses to show that they recognise the authority of the head to punish them, at least on the surface of their discourse, and that they do not contest this right. The ritualised nature of these questions can be seen when E answers his own questions: →

60E: 61E:

if it happened tomorrow do you think I’m going to believe you? no(.) would you blame me? no(.)

However, E’s question: ‘If it accidentally happens tomorrow what do you think I should do?’ clearly expects another answer than L’s response ‘Dunno’, in line 63. However, this non-committal answer on the part of A may simply be seen as not wanting to suggest punishments to the head. As I mentioned above, it is not necessary simply to analyse the boys’ minimal interventions as an acknowledgement of their guilt or as an affirmation of the power relations within this particular context. As Scott has argued, those in positions of powerlessness may strategically use minimal response or silence as a way of containing their anger and frustration, leaving the fuller expression of emotions for contexts where they find themselves among their equals, who will sympathise (Scott, 1990). Manke has also argued that teachers aim

132

Gender and Politeness

to control students to facilitate learning while students generally aim to evade adult control and to have ‘an interesting day’ (Manke, 1997: 4). The boys could in fact be seen to be resisting the authority of the head through simply not responding except sullenly, giving him only the barest of replies – a recognition that they feel that they have to endure a ‘telling off’ by the headteacher, but they do not recognise his right to do so. They can be seen as simply going through the motions of interacting with him, providing the barest minimum of a response to his questions. L’s response of ‘Dunno’ in line 63, within this interpretation, could be seen as a recognition that he is expected to respond but a refusal to do so in a way which would acknowledge the allocation of blame. Yet even within this interpretation, the boys’ replies cannot be seen as simply impolite and do not seem to be coded by the headteacher as impolite as he reacts towards them as if they are appropriate or expected responses:



61E: 62E: 63L: 64E: 65E: 66E:

you? no(.) would you blame me? no(.) If it accidentally happens tomorrow what do you think I should do? (2) dunno well if it accidentally happens tomorrow that you’re in any fights or any trouble I shall have to ask your mum and dad to come in and have a word right?

Rather than questioning L’s ‘Dunno’, E accepts it at face value and fleshes out in lines 64–66, the answer to this quasi-rhetorical question that he had been hoping for. Thus, this incident shows that behaviour which in other contexts would be impolite – minimal response to questions, humiliating and excessive irony – are here considered a simple part of the community-of-practice norms of dealing with this type of incident. As well as being analysed as the opposite of politeness, impoliteness has also been analysed as ‘merely pragmatic failure at politeness’ (Beebe, 1995: 154). Certain utterances can be classified as failure to express one’s self adequately; for example, when someone who had lived in my house twenty-five years before me visited the house, she exclaimed on entering one room: ‘Oh, you’ve still got the same carpets.’ Because of the tone of voice in which she said it, it was clear to me that this seemingly impolite utterance (‘you must be stingy as you’ve still got the same moth-eaten carpets that we had twenty-five years ago’) was not intended on her part as an insult, but a straightforward expression of surprise, since she recognised the

Politeness and impoliteness

133

carpets from her childhood. However, this type of utterance, where the potential interpretation of impoliteness is due to a misapprehension of the full implications of an utterance, or in the assessment of the politeness appropriate for a particular situation, is not symptomatic of most forms of impoliteness. Eelen argues that this view of impoliteness characterises a great deal of research, where impoliteness is seen as ‘the non-performance of an act, as the lack or absence of something’ (Eelen, 2001: 98). Beebe claims that rather than seeing impoliteness as a failure to be polite, rudeness should rather be seen as ‘a reflection of pragmatic competence’, that is, it should be seen as achieving certain aims in a conversation, firstly, to get power; and secondly, to give vent to negative feelings (Beebe, 1995: 154). In her analysis of examples such as the following, the interactants had clearly not miscalculated the level of appropriate politeness due in the circumstances, but had chosen to be rude. In New York, a well-built man was trying to park his car next to a pedestrian crossing, and a thin woman was trying to cross the road with her children. They argued about who had right of way; the woman finally yelled: Woman: Man: Woman: Man:

Oh, shut up, you fat pig! Go fuck yourself. Go on a diet! Go fuck yourself! (Gavis, cited in Beebe, 1995: 154)

This type of impoliteness, Beebe asserts, often results from a ‘volcanic’ loss of temper, or loss of control over one’s emotions; outright hostility seems to pervade many of the examples that she discusses. She argues that when an act is assessed as impolite, by one or all of the participants, it has serious consequences in the interaction. She also argues that ‘the idea that socially sanctioned norms of interaction are violated is central to the perception of rudeness’ (Beebe, 1995: 159).7 Thus, interactants will draw on what seem to them to be stable norms of acceptable behaviour in their assessment of impoliteness, despite the fact that individuals in fact assess these norms differently according to the community of practice within which the exchange takes place. In the context of New York, where Beebe carried out her research, there are certain types of linguistic behaviour which are not considered to be impolite by many of her

134

Gender and Politeness

interviewees, whereas in similar contexts in Britain they would have been. She gives the example of the use of the interjection ‘What’s your point?’ to try to shape what your interlocutor is telling you, implicitly cutting short their turn, and also implicitly suggesting that they have not made the relevance of their discourse sufficiently clear; she suggests that 40 percent of the respondents that she interviewed at a conference in America did not consider this rude, unless there was overlap or interruption.8 Impoliteness should not be seen simply as a violation. If we assume, as Fraser and Nolen do, that there is a conversational contract operating in Gricean terms, impoliteness will be seen as a violation of this contract; they argue that ‘to be polite is to abide by the rules of the relationship. The speaker becomes impolite just in cases where he [or she] violates one or more of the contractual rules’ (Fraser and Nolen, 1991, cited in Ide et al., 1992). However, impoliteness is not so easily classified, as very often it is not clear to either participant if someone has been polite or impolite. The notion of a conversational contract may be useful theoretically, but no one signs a contract or knows precisely what the terms of that contract are when they begin to converse. The terms of the ‘contract’ are subject to constant negotiation throughout an exchange. Impoliteness is often attributed to someone on the grounds of not having observed the socially sanctioned politeness behaviour which other participants assume would be expected in a particular situation, for example, the use of directness for requests which would normally be indirectly handled in English, or the lack of elements such as ‘please’, ‘thank you’, and ‘sorry’. These features are often explicitly taught to children, and are the subject of a great deal of nagging and resistance within middle-class families.9 By categorising such behaviour as impolite, the parents are displaying concern with breaches of social propriety, and the children could be seen to be refusing to acknowledge the social structures and hierarchies implicit in the insistence on the use of these elements. Jary argues that impoliteness and politeness are therefore to be considered fundamentally different in kind rather than simple polar opposites, since, instead of the Brown and Levinson view that ‘whenever the so-called polite forms/strategies are used then an additional layer of meaning is necessarily communicated . . . our experiences as conversationalists tells us that polite forms often go unnoticed by participants.

Politeness and impoliteness

135

Although there are cases when we do comment on the politeness of someone’s verbal behaviour, much of the time we don’t notice this aspect of it’ (Jary, 1998: 2). Thus, the omission of formal greetings or thanks may well be considered to be impolite, especially if that person is not liked, or if this is not the first time that socially sanctioned politeness norms within the particular community of practice have been breached. But impoliteness is not simply a question of the omission of formal or formulaic social politeness. Impoliteness can be considered as any type of linguistic behaviour which is assessed as intending to threaten the hearer’s face or social identity, or as transgressing the hypothesised community of practice’s norms of appropriacy. This notion that it is also the stability of the community of practice which is threatened in instances where someone is accused of impoliteness is important since very often accusations of impoliteness are concerned with problems of agreement over the assessment of the social standing of individuals in relation to one another, or the assessment of familiarity between them and thus the assessment of the appropriate level of politeness to use. Accusations of impoliteness generally signal to participants that there has been a mismatch in the assessment of status, role, or familiarity and thus perhaps also a mismatch in their assessment of their position in the constitution of the community of practice. Haverkate argues that we need not only to distinguish between polite and impolite acts, but also to constitute a category of nonpolite speech acts, which he claims consists of assertives and directives which are neutral with respect to politeness because they are simply giving information (Haverkate, 1988). However, I would argue that there are no speech acts which in their very essence are not perceived by others to be doing something in relation to politeness/impoliteness or at least to rapport-management. There may be degrees to which particular utterances can be said to contribute to the harmony or disharmony of a relationship, but there does not seem to be a neutral position for any utterance. In some cultures this pervasiveness of politeness and impoliteness is more clearly exemplified. For example, in Japanese, Matsumoto argues that ‘since any Japanese utterance conveys information about the social context, there is always the possibility that the speaker may, by the choice of an inappropriate form, offend the audience and thus

136

Gender and Politeness

embarrass him/herself. In this sense, any utterance, even a simple declarative, could be face-threatening’ (Matsumoto, 1989, cited in Spencer-Oatey, 2000: 19). Although Japanese cannot be considered to be the same as Western European languages, since the honorific system and the necessity to mark social positions appropriately in all aspects of the grammar do not apply to the same extent; yet, the signalling of awareness of one’s position in the social hierarchy operates in a similar way in English. As I have argued in other chapters, there are particular signs which may be drawn on to decide whether an utterance is polite or impolite. Impoliteness can be construed from the occurrence of a very wide range of linguistic behaviour. In some cases, it can be attributed to someone over a long period of time, where previous ‘signs’ of impoliteness are called upon to prove that someone’s utterance is impolite. Alternatively, impoliteness may be judged to have occurred in a fairly direct and clearly face-threatening way in a single utterance. Most analysts of politeness tend to focus only on the single utterance level rather than this form of impoliteness which is cumulative. Culpeper suggests that particularly in the familial context, impoliteness has ‘a tendency to escalate’ (Culpeper, 1996: 355). This is quite unlike politeness, where the level of politeness, in general, is more likely to remain fairly stable throughout a conversation. However, it must be added that impolite acts which are minimally face threatening can generally be tolerated in the normal course of events between people who work or live together. Thus, I should like to suggest that utterances which at face value seem impolite are not always face threatening. It is only when impolite acts are ‘added up’, or viewed in a cumulative way, and when it is assumed that the speaker intended to be impolite that they constitute a threat to the face of the hearer and to the community of practice. Thus, impoliteness cannot be said to be simply a question of the content or surface message of the utterance, but it is an assessment made on the basis of hypothesised intention. This ‘intention’ is constructed by drawing on a range of different types of evidence. Beebe suggests that intonation is very important here; she categorises a particular type of contemptuous intonation as the ‘You are Stupid Intonation’, where, when used with deliberate misinterpretation and contemptuous looks, the utterance can be classified by the hearer/s as impolite (Beebe, 1995: 165). However, each of these elements may

Politeness and impoliteness

137

be used to disambiguate the other element; thus, for example, if an interlocutor decides that the speaker is giving her/him a contemptuous look, they will be more likely to categorise other elements in the interaction as sarcastic, for example classifying their intonation or tone of voice as problematic. Impoliteness seems to have different consequences from politeness. As I argued in chapter 2, politeness has a variety of functions, and is assessed both positively and negatively depending on the context and community of practice. For example, politeness may be judged to be maintaining the status quo and ensuring that everyone is more or less at ease with one another; it may be judged to be ensuring the ‘smoothness’ of the interaction and avoiding conflict. Or, it may be seen as a trivial form of linguistic behaviour, associated largely with white, middle-class women, which holds up the flow of conversation and constrains the direct expression of emotion and thoughts. Impoliteness is viewed as having different and often longer term consequences and may, in extreme circumstances, lead to the breakdown of conversation and the disruption of a relationship. Indeed, group relations may suffer as a result of the perception of impoliteness between group members, and some may feel that they have to take sides. In contrast to assessments of politeness, I would argue that most judgements of impoliteness are not arrived at in the heat of the conversation, as most researchers discussing impoliteness seem to suggest, but they are mulled over, discussed with others, and future behaviour and strategy planned out in response to this overall assessment of someone’s behaviour over a period of time.10 Third parties may be approached to discuss someone’s impoliteness and generally involved in some repair work to the interaction and to the relationship, if the impoliteness is considered exceptional. Indeed, a great deal of interactional work goes into the assessment of impolite acts, involving retelling anecdotes and inviting judgements of the excessiveness of the impoliteness, in order to bolster the sense that one’s assessment of the impoliteness is justified or not. In this sense, the assessment of impoliteness needs to be seen as something which often involves the whole community of practice, and explicit discussion of the limits of acceptable and unacceptable behaviour in that particular group. As Spencer-Oatey and Jianyu argue, positive face within the Brown and Levinson model has been underspecified

138

Gender and Politeness

‘people have two fundamental desires for approval: a desire for positive evaluation in terms of personal qualities such as competence, abilities, etc., (quality face) and a desire for positive evaluation in terms of social identity, such as standing within a group, (identity face)’ (Spencer Oatey and Jianyu, 2000: 279). Thus, they stress the importance of the social aspect of impoliteness within their analysis. Take, for example, the conversational problem discussed by Spencer-Oatey and Jianyu (2000) when a Chinese delegation came to Britain to visit a factory with which their company traded. Because of perceptions of a lack of respect which it was assumed that the British company had shown to the Chinese group (by providing them with inexpensive hotel rooms, providing a conference room which did not enable members to be equally represented, and because a certain level of formality was not maintained), the Chinese group withdrew from the visit and complained about the impoliteness which they asserted had led to a breakdown in communication. Spencer Oatey and Jianyu interviewed all of the participants about these problems. The Chinese delegation drew attention to the fact that when the head of the British company gave a welcoming speech, where he acknowledged the importance of the relationship between China and Britain in trade terms, the Chinese delegation felt insulted. The Chinese sales manager said: ‘It is understandable for them to praise their own products, but by doing so they in fact made a big mistake. Why? Because you see because for a company when they haven’t got new orders for their products for several years it is a serious problem, to them, but they didn’t talk about it . . . he should have said that you have made great efforts regarding the [sale] of our products, right? And hope you continue. They should have said more in this respect’ (Spencer-Oatey and Jianyu, 2000: 281). The head of the Chinese delegation complained when he was not invited to give a speech immediately after the British managing director: ‘According to our home customs and protocol, speech is delivered on the basis of reciprocity. He has made his speech and I am expected to say something . . . Condescension was implied. In fact, I was reluctant to speak, and I had nothing to say. But I had to, to say a few words. Right for the occasion’ (Spencer-Oatey and Jianyu, 2000: 278). The British company was also offended, as the Chinese group decided to cancel the training sessions which had been planned for them and the British director complained that ‘they haven’t any ethics, [they] had

Politeness and impoliteness

139

no respect for their hosts’ (Spencer-Oatey and Jianyu, 2000: 278). Thus, during a ten-day visit, the impoliteness or lack of respect of one group to another was gradually assessed in a cumulative way and evidence was marshalled to support this view. ‘During the ten day visit it seems that, for the Chinese, the problematic events revolved primarily around their concerns over identity face, especially their status. They regarded themselves as being extremely important to the British company, and thus as having high status, and felt that the British hosts failed to acknowledge this sufficiently, and thus failed to give them the face that they deserved’ (Spencer-Oatey and Jianyu, 2000: 280). The Chinese felt that they as a group had not been accorded the respect which was appropriate to them, and the British also felt that they had been treated with disrespect as a group. This rootedness of impoliteness in assessments of what is appropriate to the behaviour of communities of practice is essential for an adequate analysis of impoliteness. To sum up, politeness and impoliteness cannot be taken to be polar opposites, since impoliteness functions in very different and context-specific ways. It is thus important not to analyse impoliteness in a decontextualised way, focusing only on what takes place in an interaction, as it is a negotiation or a testing out of what are perceived to be community-of-practice norms, and a discussion of those norms with others. Perhaps even more than politeness, judgements of impoliteness are explicitly concerned with the individual’s role within a community of practice. Judgements of impoliteness As I have stressed in previous chapters, politeness and impoliteness cannot be analysed in relation to single utterances or speech acts, and they cannot be assessed in relation only to what we assume are the speakers’ intentions. Impoliteness can only be understood and analysed pragmatically when considered in relation to group/ community understandings of utterances, and also in terms of the long-term discourse strategies of the interlocutors. In this section I analyse swearing and directness to examine the factors involved in our judgement of utterances as impolite. I then move on to an analysis of the factors which contribute to utterances being considered impolite.

140

Gender and Politeness

We must resist the notion that there are objective levels of impoliteness, even though interactants often behave as if those levels exist; for, as Eelen remarks, ‘In everyday interaction, judgements of (im)politeness are passed as if there do indeed exist absolutely valid norms (shared by the whole community) that form the grounds on which judgements are made’ (Eelen, 2001: 189). However, this assumption that the behaviour of others is judged according to the standards of any reasonable person – an idealised representative of the wider community values – falls foul of the type of bias experienced with sexual harassment legislation, where that ‘reasonable person’ is often interpreted as being someone who shares the values and expectations of those who make up the dominant members of tribunals (Erlich, 1999; Erlich and King, 1996). It is quite clear in many cases that theorists of politeness are themselves making moral judgements about impoliteness. For example, Blum-Kulka, in her analysis of Israeli linguistic behaviour, asserts that even when the participants do not regard their own behaviour as polite, she asserts that it is (Blum-Kulka, 1992). Ide et al. also seem to elide their position with that of the dominant classes within Japan when they describe certain highly valued linguistic politeness strategies as due to a concern to ‘beautify’ the language (Ide et al., 1992). In response, Eelen suggests that ‘if ordinary speakers invoke norms in their explications of politeness [and impoliteness], then we should not simply do the same, but rather zoom in on that activity of norm-invoking and examine it more closely, as it is likely to give us an insight into what (im)politeness actually involves’ (Eelen, 2001: 252). In addition, Eelen argues that rather than simply analysing the judgement of impoliteness, and its effects on the speaker, we should also analyse the effect that the accusation of impoliteness may have on the hearer. He suggests that the accusation of impoliteness can be used by the hearer to create a positive self-image for themselves and to put them in a morally superior position. I would like to analyse two linguistic features which are often focused on in the discussion of impoliteness: swearing and directness. Within certain environments, swearing is tolerated to a greater degree than in others and indeed can be thought of as a way of indicating one’s affiliation to others. In a survey conducted at Sheffield Hallam University in 2000 on undergraduate attitudes to swearing, it was found that whilst in their own informal linguistic behaviour

Politeness and impoliteness

141

students often considered swearing to be a way of creating informality, there was conflict over whether they disapproved of swearing in public situations such as radio interviews, television, and politicians’ speeches. Most of the undergraduates disapproved of children swearing and stated that they would find it difficult to use certain swear words themselves in front of their parents or grandparents. Most of the students made clear distinctions between the use of mild and extreme swear words, some of which they would not use themselves at all, and some of which they would use in certain contexts. There did not seem to be any major distinctions between the opinions of male and female students, although the male students seemed to be less tolerant of swearing by women. Stereotypical beliefs about the behaviour of children and women seemed to inform a great deal of the responses. Whilst these results are not surprising, they highlight the fact that undergraduates are very aware that the speech context and the community of practice determine to a great extent the degree to which swearing is considered acceptable or offensive. Whilst most people would classify swearing as impolite behaviour, in fact, there is a range of different contexts which demand different levels of swearing. Swearing does not function in one particular way: as Harris as noted, perceptions that levels of ‘bad language’ in public life are increasing may be seen by some as indicative of an increasing breakdown in the fabric of society; by others that the distinctions between the educated and uneducated are being eroded; and by yet others as a growth in a welcome informality (Harris, 1990: 417). Directness is often characterised in the literature as intrinsically face threatening. However, this is not always the case; in certain groups, especially in the business environment, a certain level of directness is tolerated, which might in other contexts be considered impolite. Miller remarks upon the fact that in a work environment which she studied: ‘the exchange of negative assessments may seamlessly unfold without participants becoming miffed or uncomfortable’ (Miller, 2000: 253). Stalpers argues that business talk is considered by many to be ‘maximally efficient communication’ and ‘business talk allows for a larger degree of tolerance for behaviour which, in other contexts, would more easily be considered impolite’ (Stalpers, 1992: 219, 220). Because in many businesses the principle that organises everything is that time is money, there is a

142

Gender and Politeness

perception that a certain amount of impoliteness is tolerated in order to get things done: ‘this principle is at variance with the general expectation of polite conduct in discourse as it is assumed for casual conversation’ (Stalpers, 1992: 230).11 There is an assumption that in English indirectness is polite and directness must therefore be considered impolite. When referring to impoliteness, some of the people whom I interviewed remarked upon the way that certain friends or acquaintances of theirs used language which they found overly direct. However, when talking about the politeness strategies of a mutual male friend [W] with another male friend [Y] who sometimes works for him, Y said, when I suggested that some people find W very impolite: ‘Yeh, but I like that sort of straightforward, no messing . . . you know where you are with W . . . he says exactly what he means. He rings up and says “There’s a problem”, tells you the problem, and then says when can you come and fix it.’ This seems to suggest that there is a hypothesised stereotype of masculinity associated with seeming directness, which can be judged by many people to be impolite in certain contexts, but which may also be seen in positive terms by those who affiliate themselves with stereotypes of masculinity. However, as I have noted before, this view of directness is based on a stereotype of masculinity, and thus, since individuals tend to hypothesise their own version of the stereotype which they feel is ‘in play’, helps us to explain how it is that different individuals will work to different stereotypes of masculinity and have different ‘takes’ on how they view the hypothesised stereotype (see chapters 4 and 5). There are other cultural stereotypes connected with directness, for often it is assumed that Chinese and Japanese people are very indirect; however, as Storti says ‘The notorious indirectness of Asians, may to a certain extent be nothing more than our inability to recognise Asian-style directness when we see it’ (Storti, cited in Harris Bond et al., 2000: 48, 49). Thus, we should not assume that a declarative or order is the only way that directness can be expressed, simply because that is the way it is often expressed in English. Miller argues that stereotypes of directness and indirectness are often drawn upon only in times of crisis; thus, in general, within a work environment, when Japanese and American colleagues interact, there is a great deal of give and take but ‘when there are misunderstandings, folk theory and popular stereotypes would lead us to blame Clint Eastwood

Politeness and impoliteness

143

style Americans who blast their way through every conversation, or compromisingly ambiguous Japanese who produce a trail of uncertainty in their wakes’ (Miller, 2000: 253). Thus, it is only when there is conflict that stereotypes about speech styles are brought into play. If we examine examples of directness in conversation, especially where those perceptions of directness have led to assessments of impoliteness being made, we can see that the judgement of directness is a subjective one, rather than directness being a feature of an act or strategy. Pavlidou argues that directness in German cultures should not be considered impolite at all; in fact, it should be seen as a way of expressing closeness and affiliation: ‘there are numerous ways of attending to the relationship aspect of communication, e.g. phatic communication, redundancy, negative politeness, talk about the relationship itself, and also strategies of directness which may result in the omission of all of the previous strategies’ (Pavlidou, 2000: 138). However, directness, whilst it may be intended to communicate closeness to another may be misinterpreted, as House has shown in her analysis of American and German misunderstandings, and Gunthner has in her analysis of Chinese–German misunderstandings (House, 2000; Gunthner, 2000). Gunthner discusses a conversation about sex-equality between two female German students, Doris and Andrea, and two Chinese students, Tan (female) and Yang (male). The conversation becomes quite heated, when Yang [Y] attempts to defend himself against accusations of sexism (as if he were being forced to defend the whole Chinese nation) and both of the Chinese participants try a range of strategies to change the topic. For example when Doris [D] says: D:

Y: D:

well I don’t quite understand why you say that eh in China there is no women’s problem the problem actually is the same it’s just that ehm that it is HUSHED up much more (quiet) not as bad as here YES THAT’S BECAUSE WOMEN ARE MORE CONSCIOUS HERE

In interviews with Gunthner after the interactions, the Chinese participants stated that they found the Germans ‘direct’, ‘aggressive’, and also ‘rude, yes a bit offensive’, whereas the Germans found the Chinese ‘boring conversationalists’ and ‘just not interesting’ (Gunthner, 2000: 218). This direct argumentative style is considered

144

Gender and Politeness

by the Chinese to be ‘very rude and inconsiderate behaviour’, whereas the German participants considered directness as showing involvement in the discussion and attention to the other (Gunthner, 2000: 237). Thus, directness should not be seen as always impolite; in fact, for these German participants, directness is seen to be positively polite. However, Chinese people in Gunthner’s survey can be viewed as equally direct, for example, when they asked questions about marital status and one’s salary, which German participants found disconcerting and embarrassing, but which within Chinese culture would be seen as appropriate small talk. Assessment of directness is a value judgement: if it is seen to be a positive feature, it is classified as positive politeness; or if negatively assessed, as impolite. One of the factors in the assessment of utterances as constituting impoliteness is the level of sincerity which we attribute to others. Jary argues that simply using a form of higher esteem for your addressee than s/he deserves (however problematic that notion is) is not enough to be polite: it must be (or appear) sincere; that is the addressee must believe that the evidence you provide actually reflects the esteem in which you hold him[/her]. [S/]He must not think that your behaviour is motivated by an attempt to raise, however indirectly, your own standing in his[her] eyes by insincerely indicating that you hold him [/her] in higher esteem than is in fact the case. Thus similar implications may be beneficial in one context (where sincerity is assumed) but detrimental in another (where it is not). (Jary, 1998: 12)

Instead of viewing impoliteness to be intrinsic to an utterance we need to see the role that context and assumptions about intentionality play. Our assessment of the context and the relation between the utterance or stretch of talk in terms of appropriateness is crucial in deciding whether an utterance or stretch of speech is impolite or not. Fraser and Nolen argue that ‘we often take certain expressions to be impolite, but it is not the expressions themselves but the conditions under which they are used that determine the judgement of politeness’ (Fraser and Nolen, 1981, cited in Haverkate, 1988: 400). For example, in situations of conflict, any expression can be interpreted as impolite. Impoliteness can be attributed to someone when we can assume that there is an intention or motivation to threaten the person’s face, as I mentioned earlier in this chapter: ‘the description

Politeness and impoliteness

145

of any aspect of a speaker’s verbal behaviour as polite or impolite depends firstly on his noting this aspect of her behaviour and secondly on the motivation he believes to be behind it’ (Jary, 1998: 8). Thus, impoliteness is about the attribution of motivations to another person. Jary argues that ‘it would be hard to maintain that the motivation for engaging in what is generally called polite behaviour is to communicate politeness. For a start, it has been argued that (im)politeness is only noted – is only relevant – when some aspect of the speaker’s behaviour provides evidence for the hearer that she holds him in higher or lower regard than he had assumed’ (Jary, 1998: 11). Impoliteness can be attributed to someone because of a mismatch of expectations or because of a cross-cultural difference in the functions or meanings of certain politeness strategies. For example, as Bargiela et al. have shown, involvement strategies used by British and American people in naming others may be understood as offensive and over-familiar by those from different cultures (Bargiela et al., 2002). These strategies of politeness can become so embedded within a culture that they appear as common sense to the members of that culture. Thus, in interactions with strangers, deference is expected by Chinese speakers and the use of the appropriate respectful name is expected, whereas for Westerners, very often, the length of time it takes to get onto first-name terms is an indication of how well the conversation is progressing. Westerners who use first names with Chinese people may intend familiarity and politeness, but in fact they communicate disrespect.12 There is an Anglocentric assumption that seeming egalitarian strategies of politeness are necessarily better than hierarchical deference strategies. However, we may question this simplistic notion of cultural difference as if all members of a culture necessarily recognise the same language norms in relation to politeness, simply because they speak the same language. There is a tendency in the literature on politeness to homogenise cultures, assuming that all members share the same values and views with regard to politeness. Particularly in relation to a culture such as Japan, where the speech behaviour of young Japanese people, especially women, differs markedly from that of older people, it is difficult to characterise the speech patterns of the whole community (Okamoto, 1995). Furthermore, as Ide et al. (1992) have shown, in the use of honorifics the majority of the population feel that,

146

Gender and Politeness

whilst they are important, they themselves do not use them correctly. It is thus essential that we recognise variation within cultural groups. A further factor in judging whether an utterance is impolite is the degree with which one is familiar with that person and the amount of investment that one has in the relationship. Culpeper has remarked that there may be a difference in the strategies of impoliteness which are employed by intimates and by strangers (Culpeper, 1996). In private conversations, it may be the case that emotional feelings are more likely to be evoked by assessments of rudeness than where the rudeness occurs in an institutional setting, that is where its use is sanctioned, in a court or in army training, for example. If you know the person well, you may be more willing to assume that the person did not actually intend to be impolite to you. When we like someone or feel that we have a stable relationship with them, we are more likely to try to excuse behaviour on the grounds of a problem with expression. However, within an institutional framework we may not be able to complain of impoliteness. Keinpointner also draws attention to the way that impoliteness may be used as a way of maintaining the distinctions between out-groups and ingroups (Keinpointner, 1997). Thus, although there is no consensus on whether familiarity with someone enables one to be more or less polite, it is clear that affect is intricately related to judgements of politeness and impoliteness. When analysing impoliteness, theorists have generally tried to analyse utterances and interactions in terms of threats to the hearer’s face and have not examined questions of emotional response to the hearer. In analysing conversations and anecdotes, it is clear to me that whether the speaker and hearer or members of a group classify themselves as liking someone or not is inextricably linked to their assessments of them as impolite. There is no simple oneway traffic between these assessments, so that it is not possible to assert that A does not like B and therefore is likely to judge B to be impolite whatever utterances B produces. Rather, A’s and B’s lack of esteem for each other is likely to develop at the same time (possibly both as a result and as a cause) as their assessments of each other’s impoliteness towards each other. As House remarks: ‘an emotional reaction is often the major factor responsible for a deterioration of rapport and for the mutual attribution of negative personal traits’

Politeness and impoliteness

147

(House, 2000: 147). In House’s analysis of conversations between German and American students, she notices that a misunderstanding leads to both of the participants deciding that they dislike each other. In the following extract, Norman [N], the American student, and Hannes [H], the German student whom Norman has invited round for a meal, converse in German (translation by House): N: H:

Hallo Hannes (0.1) good to see you (0.2) how are things with you? Oh hallo N (.) oh man well (0.2) to tell the TRUTH (.) I am very hungry what have you COOKED? (0.3) N: spaghetti? H: yeah great. So yeh but yeah I hope it’s not beef the thing is N: I well I hope you are not disappointed I have cooked SPAGHETTI and the sauce with it of course I mean I have. [not specially] H: what I’m getting at is is well (.) we should KNOW that it is from Argentine or? (House, 2000: 159) EXAMPLE 8

House suggests that in German, this type of direct strategy of getting straight into a discussion is something which signals that you consider your interlocutor to be an equal who is able to engage in an intellectual discussion; but, for Norman, Hannes ‘acted like a stranger’ and ‘insisted a bit too much and too long’ in pursuing this topic of conversation which constituted a negative assessment of the food which Norman was offering him. Hannes, in an interview with House, seemed unaware of the fact that Norman had become more and more alienated by the conversation to the point where he stopped talking. A similar pattern can be seen in another conversation which House analyses, between Brian [B], an American student, and Andi [A], a German student who has been invited for a meal. The conversation takes place in German (translation by House): B: A: B: A: B: A: B: B:

hallo Andi how are you? yeah fine oh fine really yeah so everything’s ready now (.) I hope you like it (0.3) I have cooked it myself [so because] [yeah fine] that’s what we eat in the South [loud voice] // but that’s so much that’s FAR TOO MUCH rice that doesn’t matter (0.1) I have paid for it (.) and I have INVITED you (.) [you have]

148

Gender and Politeness

A:

[no it] DOES matter it DOES it DOES think of the many poor people who go hungry and would like to eat something like that B: [well I I I] believe (0.1) I find A: I find one should in this common world in which we do all live (0.2) the world in which we are all endowed with material goods so unequally we should at least try on a small scale try to reduce no waste no useless waste. (House, 2000: 154, 155) EXAMPLE 9

In an interview with House, Brian remarked that he felt that he had been unpleasantly ‘talked at’ by his friend, whom he considered had acted like a teacher towards him. He said that he felt sad and disappointed and in his own words ‘took a dislike’ to Andi, to whom he imputed inconsiderateness and ‘selfishness’, even though, as House remarks, Andi had helped Brian with his German essays in the past (House, 2000: 156). Andi, on the other hand, did not recognise that anything untoward had occurred and simply assumed that Brian had been uninterested in the topic of conversation. There is a cumulative process at work in deciding whether you like someone or not, exemplified in these analyses above, so that if you decide that someone has been impolite to you, you may well decide that you do not like them. Impoliteness is generally considered as an orientation away from an interlocutor, and therefore a decision not to like that person is justified, because of their behaviour towards you. Spencer-Oatey discusses orientation to others in the following terms: we may have an orientation to another which can be classified as rapport-enhancement (concerned to improve rapport); rapport maintenance (concerned to keep the rapport as it is); rapport neglect (where no efforts are made to improve rapport), and rapport challenge (where face threatening acts are committed and it is clear to the parties that the relationship is under threat) (Spencer-Oatey, 2000). However, whilst we can recognise that these categories are useful in deciding whether someone is positively oriented to us or not, these orientations are not clearly signalled to others; they are something which participants constantly assess and make decisions about whether they consider particular types of behaviour as evidence that the other wants to enhance rapport with them or not. Thus, when analysing impoliteness, we are also making judgements about whether we are oriented to someone or not.

Politeness and impoliteness

149

Thus, there is a range of factors which contributes to our judging an utterance or exchange as impolite, some of them based on stereotypes of gender or culture and some of them based on assessments of sincerity, speaker intention, and appropriateness. Class, race, and impoliteness Factors of gender, class, and race are not generally considered when analysing politeness and impoliteness, but there is a stereotypical assumption that white, middle-class people, particularly women, are more polite than other groups. I shall be discussing the issue of gender in more detail in chapters 4 and 5, but here I should like to discuss racial and class stereotypes of politeness and impoliteness (Mills, 2002). Representations of working-class people tend to stress the directness and loudness of their language, and emphasise the degree of swearing that is used (see, for example, representations of the working classes in television programmes such as the Royle family, on working-class life in Manchester, and in novels such as James Kelman’s on Glaswegian working-class life). Because politeness is often associated with ‘civility, ‘courtesy’, ‘good manners’, ‘good breeding’, and ‘a good upbringing’, all qualities associated stereotypically with the white, upper and middle classes, it is not surprising therefore that working-class people and Black people are characterised as impolite. There are stereotypes of race which are similar to those of working-class behaviour, particularly in relation to directness. When Labov discussed the ritual insulting or ‘soundings’ of New York Black young men, it was analysed as if it were ‘mock impoliteness’ and, when compared with stereotypes of white, middle-class politeness norms, it was judged to be aberrant (Labov, 1972). Eelen argues that ‘although labelling such behaviour “impoliteness” may arguably not constitute a moral condemnation by the researcher (indeed it is mock impoliteness) it clearly does represent a morally involved point of reference’ (Eelen, 2001: 181). We could perhaps argue that this particular community of practice had taken stereotypes of white, middle-class speech and openly flouted them or played with them as a form of positive politeness, thus forming themselves as an out-group through their use of an anti-language. However, we cannot generalise the behaviour of this very small group of males to make statements about politeness and

150

Gender and Politeness

race except at the stereotypical level. In fact, Morgan argues that Black/African American women have been marginalised in much of the work on Black American signifying, and that some aspects of their verbal play can be seen to bear striking similarities to that of adolescent males, particularly when they are engaged in competitive activities (Morgan, 1999: 33; see also Henley, 1995). Morgan also draws attention to the practice of instigating, where black female interactants are assumed to have talked behind someone’s back: ‘instigating events are therefore about participants and occurrences of talk, as well as about what was allegedly said by whom. For teenagers the event is designed to expose and either acquit or convict the instigator and the offending party . . . the offended party’s aim is to determine who started the rumour. In the process, friendships are tested, conversational roles are assessed, and all parties become invested in identifying the alleged perpetrator of the offending speech event’ (Morgan, 1999: 37). Harness Goodwin has also analysed the way that African American girls and groups of girls from a variety of ethnic backgrounds negotiate their roles and positions within communities of practice in games in the playground (Harness Goodwin, 1998, 2001). Particularly in her analysis of girls playing at groupskipping, or jump rope, she shows that in an activity which the girls have skill in, their levels of directness towards boys and other outgroup members whom they wish to exclude, is fairly high. Thus, in all of these different studies, these assessments of impoliteness in signifying, instigating, and in directives, whilst perhaps taking different forms from that stereotypically associated with white, middle-class speech, seem to perform the same function. This is not to assume that all working-class people and black people behave in different ways linguistically from the white, middle-classes, or even that their politeness norms are necessarily different. However, at a stereotypical level, working-class people and Black people are often discussed as if this were indeed the case. Furthermore, we need to consider whether the language of black and working-class groups is homogeneous, for we need only to consider the differences in the way that working-class people speak when interacting with members of their own community and when they speak to those they consider to belong to out-groups (Trudgill and Chambers, 1991). Whilst it is clear that in certain contexts, such as in competitive talk, informal affiliative talk, and in talk organising play, the language usage of

Politeness and impoliteness

151

both black and working-class people can be seen to be different in relation to the politeness norms of white, middle-class people, it is debatable whether this can be said of their linguistic behaviour as a whole. Politeness, and particularly notions of etiquette, have played an important role in distinguishing between and keeping separate the working and middle classes. As I noted in chapter 2, ‘historically a number of different factors seem to be involved in determining politeness: aspects of social hierarchy (the court), and social status (life in the city), but also a more general notion of “proper behavioural conduct”’ (Eelen, 2001: 1). This notion of ‘proper’ behaviour is important as all other forms of behaviour are then judged to be aberrant. This is particularly interesting in relation to etiquette, where clear guidelines are given for correct behaviour in a range of public formal settings, for example, weddings, balls, and formal dinners (Bolton and Bolton, n.d.). Because of changes in the level of informality within the public sphere, and increased social mobility, it is assumed that such guides to ‘proper’ behaviour are only needed in particular circumstances where formal rules still apply (Fairclough, 1992). For example, in the 1970s Bolton and Bolton argue in their guide to etiquette that: ‘since the last war, due to improved and universal education, higher salaries and greater equality of opportunities, there has been a great “levelling up” and many of the erstwhile class barriers, once so sharply defined, are gradually being broken down so that probably the only noticeable distinctions left today are in manners, deportment and speech’ (Bolton and Bolton, n.d.:13). It is assumed by them that the language behaviour aspired to is that of the aristocracy or middle classes. Etiquette books are far less common at present; however, even outdated notions of U and non-U behaviour still prevail and are clearly concerned with distinguishing between middle and working classes. Etiquette thus refers to a form of behaviour which is seen to be important in maintaining a clear distinction between classes, and it is a form of aspirational behaviour for those in the working classes or lower middle classes who would like to appear to be more middle class. However, we should not assume, as many who value etiquette do, that these politeness norms are fixed and invariable, since there have been major changes in what is considered appropriate linguistic behaviour at formal occasions, and these stable positions should be viewed as an attempt

152

Gender and Politeness

by those in the dominant classes to demarcate the boundaries of their class. Eelen argues that we should see social struggle at work in all assessments of politeness and impoliteness; by social struggle he means ‘not only . . . large scale social phenomena such as political or class struggle, but also . . . very small scale phenomena such as interpersonal differences of opinion – especially about moral issues of right and wrong, good or bad, etc.’ (Eelen, 2001: 227). Whilst I would agree that social struggle is enacted at the interpersonal level, I feel that the wider class struggle should not be lost sight of, as it often is in Eelen’s account.13 Skeggs suggests that ‘there was a time when [the concept of class] was considered necessary by the middle classes to maintain and consolidate differences in power; its recent invisibility suggests that these differences are now institutionalised’ (Skeggs, 1997: 7). This statement is important in pointing out that descriptions of someone’s behaviour as ill-mannered or impolite may often have a social function, in that the person may be classified on the grounds of stereotypes of class-appropriate behaviour. Judgements of politeness and impoliteness are often related to normative views of social class, and the models which many theorists have developed to describe politeness run the risk of consolidating the status quo; in these analyses ‘politeness becomes a set of norms, ensuring that people behave according to their place in the social structure . . . so politeness has a functional role in relation to the structural maintenance of society. It ensures its internal coherence, it ensures that everybody knows his or her place and remains within its confines . . . The politeness norms make sure that those in powerful positions receive due respect and deference, while the powerless behave in appropriately powerless ways, so the system is involved in making sure that those in power remain in power’ (Eelen, 2001: 200, 201). This analysis rather smacks of conspiracy theory, where Eelen presents a view of individuals simply producing the language appropriate to their class position in politeness, which does not accord at all with the way that people actually behave. Perhaps Eelen also only focuses on the behaviour of some workingclass people when in interaction with middle-class people, and does not consider middle-class and working-class behaviour when with members of their own communities. Thus, politeness is associated at a stereotypical level with white, middle-class behaviour and many

Politeness and impoliteness

153

researchers in this area assume that other groups are deviant in relation to this norm, and that class and racial position is simply reflected in speech rather than being negotiated by participants. Analysis of incidents judged to be impolite I would like to analyse a number of incidents which have happened to me or to others which have been judged to be impolite. The first is an example from an interview where a young white, middle-class man described an incident in which he had found a stray dog outside his house. He took the dog in, although he had several dogs of his own, and rang the telephone number which he found on the dogtag. When he told the person who answered that he had found her dog, she did not offer to collect it immediately, and instead started to tell him how difficult it would be for her to collect the dog, as she was looking after three children. He commented in the interview that he did not feel as if he had been adequately thanked for finding the dog. Thus, he assessed the level of politeness which he had been accorded as incommensurate with the effort and hence the general ‘civic duty’ which he had expended in keeping the dog. An incident that I classified as impolite happened to me when, at the beginning of a colleague’s lecture which I was attending, a group of workmen started drilling into the wall just outside the door of the lecture theatre. I went over to them and said in a manner which I thought was considerate and polite, ‘Do you think you could possibly stop drilling for a while as the lecture is just about to start?’ The person who was drilling did not turn round to talk to me, but simply said ‘OK, love’ with very flat intonation, and carried on drilling. He did eventually stop, but I assessed his behaviour as impolite, because he had given me no eye contact, and because he had not given me the sort of reply that I considered commensurate with the amount of politeness that I had used in my request. Calling me ‘love’ seemed to me to be calling attention to my gender in a particularly patronising stereotypical way, when I did not consider gender to be a relevant factor in the exchange. To me, it seemed to be saying ‘what right have you to ask me to do anything, you’re just a woman’, whereas my request had, I thought, been saying to him, ‘I am a lecturer here and I have a right to ask you to stop, but I am prepared to do it in a considerate way, with an appropriate level of

154

Gender and Politeness

respect for you.’ Particularly since this was a very public context with about a hundred students waiting in the lecture theatre, listening to the interaction, this did seem to be an incident where conflict over positions of power and who could ask whom to do what were being contested.14 I should like to focus in more detail on an incident which occurred at a university departmental party and which involved myself, a female postgraduate, and a new male member of staff.15 This analysis is intended to demonstrate that stereotypes of gender play an important role in assessments of impoliteness. The way that gender works in each interaction may differ markedly from the way it operates here. Focusing on an interaction where different views of what actually happens is complicated, but it illustrates some of the difficulties in assigning clear values to elements within a conversation in relation to impoliteness.16 A departmental party is a community of practice with different norms from those in the work environment; it is a complex and sometimes rather tense environment, where the interpersonal and institutional relations between staff in a department are played out and negotiated. Linguistic behaviour which might be considered impolite within the office or teaching situation, when uttered at a staff party may be considered differently. A departmental party is usually an arena where a certain amount of banter between social equals occurs; and this type of public verbal play seems to be coded by many women as a masculine way of interacting, but which females may also engage in (Yedes, 1996; Labov, 1972).17 However, as Walsh has shown in her analysis of women in the public sphere, such as priests, MPs, and campaigners, women often use styles of speech in their interventions in the public sphere which are coded as masculine, but they run the risk of being judged as transgressive or abnormal for engaging in them (Walsh, 2001; Liladhar, 2001). In the incident in question, a departmental party had been thrown to welcome a new male member of staff who was a poet. He had not been introduced either to myself or the postgraduate. This person, like us, is white and middle class and probably roughly the same age as myself, but older than the postgraduate. When he approached us, the postgraduate and I, who had been talking together, tried to be positively polite and friendly by saying ‘Hi there’ and asking the person how he was. Since the party was well underway, I felt I

Politeness and impoliteness

155

had to think up some form of appropriate phatic communion.18 I considered that banter was not an option, since I did not know the person; however, it should be noted that that assessment may be to do with the stereotypical coding of banter as masculine. Since this person is a poet, I asked: ‘What sort of poetry do you write?’ to which he replied, ‘Name me six poets.’

This response on his part confused me. If I wished to continue to classify what we were engaging in as polite small talk, then I would have to comply and provide a list of poets. I would thus have to assume that there was a longer term relevance to his request for the names of six poets which would become apparent as the conversation unfolded. However, I did not wish to be forced to answer this question, which I felt was offensive and which I glossed as his attempt to state that he would not talk about his writing, as he assumed that I knew nothing about poetry. Under this interpretation, he was in fact implying that I could not name six poets. Proxemic cues, such as body stance, eye contact, facial expression, and his intonation and tone of voice, all led me to interpret the relevance of his statement to my question as impolite. What has since become clear is that the male staff member was extremely anxious about the departmental party, and had inferred that my intended positive politeness towards him, because he considered it to be excessive, was in fact patronising and therefore insincere, and impolite.19 A further interpretation which I have only come to recently is that this conflict developed precisely because of gender stereotyping: here, a famous male poet found himself in conversation with a female professor in his department and she started the conversation with a gambit which showed that she had never heard of him. His aggression and impoliteness may have stemmed from this difficulty in accepting a relatively powerless position where gender was enmeshed with power difference (see Cameron, 1998a). I would argue that gender played a part in our attempts at making sense of each other’s seemingly inexplicable interventions. As Cameron states: gender is potentially relevant (to understanding conflict-talk) to the extent that it affects the context-specific assumptions that the man and the woman bring to bear on the work of interpreting one another’s utterances. If there is a divergence of interpretation between the parties . . . a satisfactory

156

Gender and Politeness

explanation must be sought not in gender-preferential responses to a particular linguistic strategy, but at the level of assumptions and inferences which are specific to the situations these conversationalists find themselves in. (Cameron, 1998a: 448)

In this case, the conflict seems to involve the assessments each of us made as to the level and sincerity of politeness, on the one hand, and to the overall relevance of the utterance to the conversation as a whole, on the other. These assessments and interpretations of the interaction are inflected with hypothesised gender stereotyping and assumptions. At this point in the conversation, the female postgraduate, who had been standing next to me and who had seen that I was having difficulty with the conversation, rejoined the conversation and we both attempted to try to change the subject and to resolve the difficulty. However, the male staff member then made comments which we both considered clearly impolite, consisting of overtly sexual comments and verbal aggression. Rather than simple banter which plays around with what is acceptable, sometimes overstepping the bounds of acceptability for the purposes of humour and camaraderie, this incident did not feel as if it could be classified as banter and therefore positively polite, but instead had to be classified as offensive and impolite.20 What is also important is that the male member of staff was behaving in a stereotypically masculine fashion, drawing attention to our femaleness and sexuality. This felt like aggression and not banter primarily because we did not know him. If this behaviour had come from one of our male colleagues with whom we felt at ease, we should not necessarily have considered the incident impolite, but maybe have excused it on the grounds of drunkenness and personal style more readily. As it was, neither the postgraduate nor I responded with what we considered impoliteness, but continued to use positive politeness strategies, suggesting that we talk on other subjects, or explicitly drawing attention to the fact that we seemed to be misunderstanding one another, perhaps stereotypically ‘feminine’ responses to what we saw as threatening behaviour. Because of these strategies we were locked into the interaction; we could not simply walk away. We tried to assuage him and calm him down, partly because we did not want the incident to escalate and ruin the party. Thus, all of the participants in this interaction were inferring politeness or impoliteness in relation to norms which they

Politeness and impoliteness

157

thought existed within that particular community of practice, and these norms, I would argue, have something to do with gendered domains and stereotypes of gendered behaviour (which I discuss in more detail in chapters 4 and 5). An initial coding of an utterance as impolite or polite led to a range of different behaviours for each participant. For myself and the postgraduate, it led to a range of ‘repair’ behaviours, a stereotypically feminine response, perhaps, whereas for the male staff member, it led to an increase in insulting terms, as if perhaps these were implicit from the beginning.21 One could argue that this person gained some interactional power through this type of behaviour, since he had insulted a person who was senior to himself in institutional terms (and, in fact, my status was something which was brought up later in the interaction) and also had insulted someone to whom he should have had some responsibility since she was a postgraduate student within the department. However, we need to be careful about the elision of interactional power with masculinist stereotypical behaviour, which in many contexts such as this one, does not necessarily bring any form of power to one’s self. The question of a person’s commitment to a particular speech act is important here. Walsh has argued that we need to be able to discuss the notion of inferred sympathy or commitment which we assume is behind a particular speech act (Walsh, 2001). The postgraduate and I, as participants in a particular community of practice, inferred a certain degree of commitment to this person’s speech acts. What is interesting is that those who tried to help to resolve the problem suggested that we should not attribute commitment by him to his utterances on lines which seemed strikingly gendered; that is, he is a poet (and presumably male poets have a certain type of behaviour which is seen to be acceptable), and that he was drunk and therefore should not be held responsible and committed to what he said.22 Further gendered stereotypes were brought in, since we were told that we should simply accept this behaviour because ‘that’s just the way he is’. Having seen the way that impolite, ‘masculinist’, females are dealt with in a work environment, it is worth considering the very different ways in which females are judged for directness and verbal aggression. Thus, this impolite behaviour was judged to be not serious or problematic because those who were trying to resolve or minimise the difficulty, for the best of motives, that is, in

158

Gender and Politeness

the interests of departmental harmony, were drawing on gendered stereotypes of what was appropriate behaviour for men and women. He did not mean it – he was just behaving as men do, and we should not make a fuss, as women often do. What is important in terms of the analysis of impoliteness is the outcome of this behaviour, where all of the people who attended, and some of the rest of the department, were drawn into various behaviours which either tried to resolve or worsen the perceived breach. Several male and female members of the department refused pointedly to speak to the new member of staff; several meetings were held between senior staff and the postgraduate, where she tried to make a formal complaint.23 After several weeks of not communicating with the person, I decided to try to resolve the matter by talking to him explicitly about the event and suggesting that we begin to speak to each other again. Generally, I would characterise both myself and the postgraduate as strong speakers who are confident in the public sphere. Thus, this may seem to be a fairly stereotypical feminine response to the situation, or even perhaps an admission of some fault on our part.24 However, resolving breakdowns of communication seems to me a fairly powerful move to make, and strategic use of stereotypical gendered behaviour cannot be considered in the same way as other less foregrounded gendered behaviour. This type of strategic use of stereotypical behaviour requires us to analyse more carefully the notion of the meaning of such behaviour. I considered that the impoliteness which we had judged to have taken place was beginning to reflect more on us than it did on him; I did not wish to be cast in the role of victim and he showed no awareness of the distress his verbal attack had caused, particularly to the postgraduate. This strategic use of feminine ‘co-operative’ strategies should be seen as a way in which female behaviour cannot be equated with stereotypes of behaviour, and shows that even those stereotypes can be used for our own ends. However, whilst I felt that I was resolving the situation by drawing on these feminine norms strategically, that is not to say that other members of the department or indeed the staff member himself interpreted them in this way. Thus, what the analysis of this incident shows is that gender in an interaction is not simply about the gender of the speaker or hearer. This particular community of practice is coded by many of the participants as masculine, because banter and verbal play is considered

Politeness and impoliteness

159

to be the normal mode of interaction; however, what was interpreted as impoliteness on a male’s part is condoned more, since this fits in with the stereotypes of masculine interaction. A seemingly feminine response to the situation, that is one which attempts to resolve the situation, cannot be simply coded as powerless, since in fact this is what brings the incident to a close. However, even though this is a strategic use of stereotypically feminine behaviour, it may still be classified by others as a ‘weak’ form of behaviour. Stereotypically masculine speech styles may be condoned more when they are employed by men than women, because these accord with notions of the habitual styles of men and their use of politeness. However, we should not assume that interactional power is necessarily achieved by the use of masculinist speech such as banter, directness, and impoliteness. Thus, when analysing politeness and impoliteness in relation to gender, it is not enough simply to analyse males’ and females’ use of seemingly self-evidently politeness strategies within particular interactions; what must be focused on is the perceived norms of the community of practice which lead to judgements of impoliteness being made. Conclusions It is clear that when we judge politeness and impoliteness we are also categorising people in a range of different ways in relation to what we think are cultural or community-of-practice norms. Instead of assuming that particular acts such as directness or swearing are in essence impolite, we must analyse the way that individuals come to a judgement of an utterance or series of utterances as polite or impolite. We must also be aware that individuals do not necessarily come to assess an utterance as impolite immediately, but through a process of consultation with others. Thus, what I am arguing for here is a greater complexity in the analysis of impoliteness which perhaps can only be achieved through turning from the sentence level to the level of discourse. The notion of community of practice can provide a framework for analysing the complexity of judging an utterance as polite or impolite, and by analysing individual assessments of stereotypes we can see that within different communities of practice individuals may perform their gendered, raced, and classed identities in different ways.

160

Gender and Politeness

Notes 1. It could be argued that Brown and Levinson do, in fact, consider impoliteness implicitly, as a great deal of their analysis is taken up with the description of Face Threatening Acts (FTAs); however, most of their work is concerned with the description of politeness as the avoidance of FTAs, rather than concentrating on the nature of FTAs or impoliteness. 2. However, as I discuss in ch. 4, this notion that combative styles in academic discussion are acceptable to all participants assumes that this stereotypically masculine style is in fact neutral. Many female academics and some males find this combative style very uncomfortable and unproductive of genuine debate. 3. Indeed, I would argue that we need to reconsider the self-evident nature of these speech act categories such as ‘threat’, for in order to classify something as a threat we have to take up a position in relation to the utterance and align ourselves either with the speaker or the hearer. Categorising something as a threat is an evaluation of the utterance, and an alignment with the hearer, rather than an analysis. 4. However, whilst in the normal run of events, teasing and mock insults are used to try to manage the daily working out of power relations and the allocation of tasks, it is possible that these strategies may not be successful – where a member of staff may well interpret the insults literally, not as an example of affiliation and positive politeness, but as an instance of impoliteness. 5. A participant at a workshop I conducted in Utrecht stated that when he did his year’s army training, he found the level of impoliteness personally threatening and offensive. However, whilst classifying this style of speech as impolite on a personal level, nevertheless, he recognised that it was institutionally sanctioned within that context and did not in fact complain. 6. Interestingly, Shaw (2002) shows that although there are clear regulations on the type of language that is permitted in the House of Commons and protocols for intervening in debate which are supposed to be adhered to by all, there are also procedures for informally breaching these regulations. She argues that women MPs generally adhere to the formal overt rules of interaction, but that they tend not to break the rules in the way that most male MPs do. 7. Beebe is thus making a clear distinction between accidental impoliteness and intentional rudeness, a distinction which I do not retain in this book as I would argue it is not a simple distinction but one which has to be established by the interactants themselves. 8. Within a middle-class academic context in Britain, ‘What’s your point?’ would generally be seen as a fairly aggressive form of questioning if delivered with a particular tone and loudness and with eye contact. However, there are many contexts in which this question might be considered a reasonable request for the speaker to sum up, rather than as an accusation

Politeness and impoliteness

9.

10.

11.

12.

13.

14.

15.

161

that the speaker is not being clear. It is therefore difficult to generalise about whether this would be a clearly impolite utterance. The absence or presence of such elements in the speech of children may be remarked upon by others, especially by those who consider that standards of politeness have declined in recent years, as some of my interviewees remarked. Commenting on the politeness of children may be a way of praising the parents, since for certain groups within the population, particularly white, middle-class groups and those aspiring to middle-class values and status, it may be considered to be an indicator of certain forms of disciplinary child-rearing practice, which are under threat from ‘liberal’ values and educational practices. However, this is not the only function or interpretation of comments on the social politeness of children, some of them having to do with class position and others to do with general expression of affiliation and affect. As I noted in the previous chapter, in the analysis of example 6, in the discussion between G and J where J calls on G to help him to decide whether his class acted impolitely to him or not. However this varies from context to context; large companies tend to have a different ethos from smaller companies, and it depends on what sort of product the company produces as to whether the ‘time is money’ philosophy holds sway. This question of a mismatch of cultural norms in relation to politeness does not simply affect members of different nations. Some lecturers who ask their students to call them by their first names and who call students by their first names, thus asking for reciprocity, may offend certain students by appearing too familiar. Whilst class has become a much more fluid concept in this post-Marxist era, it would be foolish to concur with assertions by Prime Ministers John Major and Margaret Thatcher that Britain is a classless society. Britain is still rigidly stratified on class lines, but other factors work in conjunction with class (see many of the essays in Joyce, 1995). However, we could consider that perhaps the workman had had a bad day and had not been able to complete his work as he had wanted, because of other lecturers asking him to stop drilling. It also might have been a reaction to a perception of over-politeness and therefore it might have been seen by him as an over-emphasis of middle-classness on my part (that is, he could have thought I was pulling rank). Added to this is the fact that in Sheffield, where this incident took place, ‘love’ has a wider distribution as a term used by men to strangers, both men and women, than it does in other parts of the country. Thus, this seemingly impolite usage may have been habitualised usage on his part. This section is a revised version of part of an essay entitled ‘Rethinking linguistic politeness/impoliteness and gender identity’, in Lia Litoselliti and Jane Sunderland’s edited collection Discourse Analysis and Gender Identity, John Benjamins, 2002.

162

Gender and Politeness

16. I should make clear that this analysis is not an attempt to ‘get back’ at the person involved, although as Eelen (2001) notes, in accusing someone of impoliteness you are necessarily establishing for yourself a position of seeming higher moral authority, whether that is intentional or not. I am interested in the aftermath of the event within a particular community of practice and what it tells us about politeness and impoliteness. Even over two years later, the incident still had effects on the department and was still occasionally referred to. There are clear difficulties in working on this material, since I am making this incident public and presenting a particular view of the event. The male member of the staff involved in the incident has received copies of this account, and I asked his permission to publish it. I also requested comments on his interpretation of the incident. He prevented me from publishing an earlier version in the department-based e-journal: English Studies: Working Papers on the Web. I should make very clear that the views expressed here about the meaning of the incident are mine alone, and I am sure that his version of what actually happened is very different and casts my actions in a very different light. However, I have included analysis of this incident simply because of what it tells us about the way people within communities of practice deal with incidents which are judged to be impolite. 17. Many of the female university lecturers to whom I have spoken about banter have stressed the fact that they see ‘doing’ banter and verbal duelling with male colleagues as a necessary but rather tedious element in their maintaining a position within the departmental hierarchy. They see it almost as a precondition of being accepted as a ‘proper’ university lecturer that they can adopt this masculinist way of speaking. It is not something that they necessarily want to do, but it is a style of speaking that many of them feel that they can use for particular effects/purposes (see Walsh, 2001). 18. If the incident had taken place earlier, it would have been possible for me to draw on a whole range of other items of small talk, such as comments about the house where the party was being held, or even the weather, but the timing of the interaction precluded the use of these. 19. It is difficult to work out what the other participant considered happened during this interaction: despite several attempts to discuss this issue with him, he has not responded. However, another member of the department who has attempted to be ‘objective’ about this interaction made comments to me which lead me to assume that this is roughly how he interpreted our actions. This may, however, be a post hoc rationalisation on that member of staff’s part, or indeed on the part of the new member of staff (just as my analysis may well be). 20. As Yedes (1996) has shown, verbal teasing and banter are only an appropriate speech style to those who know each other well, and may be misinterpreted when used between relative strangers; such verbal play may also be used strategically by those strangers who wish to

Politeness and impoliteness

21.

22.

23.

24.

163

be impolite because of this ambiguity about whether it is a signal of intimacy and therefore positive politeness or impoliteness. By this I mean that the way that the conversation developed into an excessive display of insult and sexual antagonism perhaps means that these elements of conflict were already embedded within the initial interaction where there might appear to be a certain ambiguity about whether the male member of staff intended to be polite or not. Cameron (1998a) argues that whereas Deborah Tannen (1991) considers that men and women simply misunderstand each other’s intentions, that they have different speech styles which lead to breakdowns in conversation, in fact, the conflict between men and women is one of social inequality and differential access to resources and goods within the public sphere; this is what leads to conflict. For her, conversational breakdown is seen as an instantiation of a wider conflict over power. I would agree that alcohol affects what we say to people, and when we judge that someone is drunk we also adopt different strategies towards them and judge their utterances in different ways. However, that cannot lead us to assume that the speech acts of those who are drunk should not be counted as having effects or outcomes. Furthermore, the way that drunkenness is judged as appropriate or inappropriate for men and women is striking here (see Clark, 1998, for a discussion). When the staff member was informed that the postgraduate was considering making a formal complaint, he left a note in her pigeonhole which apologised in a fairly minimal way. No formal complaint was made. This is also why I feel that it is important to see politeness and impoliteness over long stretches of interaction, because it is quite clear to me that there are several points in the interaction where the meaning of certain acts began to change their meaning for me and therefore required a different response.

4

Theorising gender

Introduction In recent years the analysis of gender in feminist linguistics and in feminist theory in general has radically changed. Before going on to analyse the complex relationship between gender, politeness, and impoliteness in chapter 5, in this chapter, I discuss the theoretical and methodological problems in feminist linguistic analysis in relation to the question of ‘women’s language’. I then consider feminist thinking which tries to move beyond the assumption that women’s speech is always necessarily different from men’s speech, and I examine the complexity of gender when analysed alongside other variables and stereotypical forms.1 Finally, I analyse the language of strong women speakers and gendered stereotypes to challenge further the notion that women’s language is homogeneous.

Dominance or difference? Feminist language research in the 1970s focused on the question of male dominance and female deference in conversation (Lakoff, 1975; Spender, 1980). It criticised both the social system which it viewed as patriarchal, and which it saw as forcing women to speak in a subservient way, but also individual males who were seen to violate the rights of their female interlocutors. Robin Lakoff’s polemical analysis of what she considered to be female language patterns was one of the first feminist linguistic analyses which made a clear causal connection between the social and political oppression of women as a group and their linguistic behaviour. This subordinated status was displayed in the language patterns which she 164

Theorising gender

165

describes as ‘talking like a lady’ (Lakoff, 1975: 10). She gives the example of two sentences which, she suggests, characterise the difference between women’s subordinated language and men’s dominant language: 1. Oh dear, you’ve put the peanut butter in the refrigerator again 2. Shit, you’ve put the peanut butter in the refrigerator again. (Lakoff, 1975: 10)

The first, Lakoff asserts, is women’s language and the second is men’s language; this distinction is made primarily on the basis of perceptions that 1 is more polite than 2, because of the ‘softer’ expletive which mitigates the force of the utterance and therefore seems to be less of a challenge to the interlocutor’s face. Lakoff makes a connection between seemingly stronger expletives and stronger positions in relation to power. As she argues: ‘if someone is allowed to show emotions, and consequently does, others may well view him as a real individual in his own right, as they could not if he never showed emotion . . . the behaviour a woman learns as “correct” prevents her from being taken seriously as an individual, and further is considered “correct” and necessary for a woman precisely because society does not consider her seriously as an individual’ (Lakoff, 1975: 11).2 Thus, within the work of early feminist linguistic theorists like Lakoff, femininity and femaleness are elided, and powerlessness is seen as a major factor in the constitution of femininity. Lakoff and Spender argued that women’s language style was further characterised by the use of elements such as hedges, tentativeness, tag-questions which seemed to these theorists to signal indirectness, mitigation, diffidence, and hesitation. In contrast to this, male speech was characterised as direct, forceful, and confident, using features such as direct, unmitigated statements and interruption. As a polemic, this early feminist research was extremely important, since it challenged the way that certain males felt that they could act because of their status within the society as a whole. It also made interruption, directness, and sexist remarks by men, and deferent submissive behaviour by women, less ‘natural’ or ‘common-sense’, and ‘just part of being a woman or man’. Thus, this consciousness-raising research, which was very widely read by people outside academic circles, made a major impact on many women, forcing them to reflect on language use as an indicator of power relations, and indeed,

166

Gender and Politeness

encouraging them to make meta-linguistic comments on language use in interaction.3 However, critics have noted that this type of analysis seemed to be focused on the stereotypical language usage of a very small group of women, that is middle-class, white, Anglo-Americans. It was not based on the examination of any data, but rather on personal anecdotes which seemed to uphold a stereotype of submissive women, without any counter-examples being considered. Many feminist linguists, such as Tannen and Coates in the 1980s and 1990s, rather than analysing dominance as such, since it was clear that the nature of power relations between women and men was being fundamentally changed at this time, turned to an analysis of the socially constructed differences between women’s and men’s language, seeing these as akin to dialects spoken by different groups who interacted with each other, rather than seeing them as simply dominant and dominated group (Tannen, 1991; Coates, 1988, 1996). This difference in female and male linguistic performance occurs, Tannen argued, because women and men are largely brought up and socialised in single-sex groups where they develop different language preferences and styles. Women and men have different aims in conversation, which lead to breakdowns in communication or misunderstandings, since women, she argues, are largely concerned to establish rapport between members of a group and to ensure that conversations go smoothly (rapport talk), whilst men are concerned to establish their place in the pecking order and use the production of information as way of establishing a position in the hierarchy (report talk).4 This focus on difference has been widely criticised by TroemelPloetz (1998) and Cameron (1998a) for its reactionary political stance, and for its failure to acknowledge the inequality that persists in many relations between women and men. What Troemel-Ploetz is most concerned about is the erasure of the factor of power difference in the analysis of interaction between women and men, and although she, as other feminist linguists, does not wish to characterise all women as subordinate, the erasing of power from feminist analysis is a worrying trend, since it implicitly assumes that men and women are equal. Cameron goes further than Troemel-Ploetz in critiquing Tannen’s work, since she argues that ‘power relations are constitutive of gender differentiation as we know it’ (Cameron,

Theorising gender

167

1998a: 438). Furthermore, Troemel-Ploetz argues that women and men do not, in fact, grow up in homogeneous and separate linguistic communities, but rather spend a great deal of their time in mixed-sex environments, whether in the school, the home, or at work.5 The positive aspects of the ‘difference’ type of feminist analysis is that it generally calls for a re-evaluation of the styles which are stereotypically associated with women; thus, Coates (1988) argues that we should re-value what has been classified as gossip and cooperative strategies/rapport talk, in general, and Holmes argues that what she claims are women’s styles of politeness should be seen as, in fact, more productive for debating issues than masculine styles of speech, as I discuss in the next chapter (Holmes, 1995). This reevaluation of women’s speech styles has made an important impact in certain areas; for example, in the evaluation of oral performance in secondary schools in Britain, it is generally those aspects of speech associated with feminine speech styles which are most highly valued in assessments of oral skill (for example, supportive comments, minimal responses, concern for others in the group, and so on) (Wareing, 1995). This is a significant shift from other ways of assessing oral performance which are more concerned to evaluate, for example, rhetorical skill and confidence. Furthermore, companies which decide to train their employees in communications skills generally tend to focus on the empathetic co-operative skills such as listening, group decision-making, and turn-taking rather than on individual rhetorical performance. Cameron has noted that, perhaps, this view that women are more co-operative than men, that their language is concerned with establishing rapport rather than with dispensing information, based as it is on stereotypes of women’s speech, has also led to the widespread employment of women in the communications industries, such as call-centres (Cameron, 2000). The so-called ‘feminine’ skills of communication, however, are not highly valued and workers in call-centres generally receive low salaries. This process of re-evaluation of what has been considered to characterise women’s speech has been of great value. However, this cannot compensate for the fact that, in general, the shift in the way that women really speak and are evaluated when they speak has been in the direction of women adopting wholesale what are seen to be masculine ways of speaking in the public sphere (and sometimes being negatively evaluated for using this type of language) (Walsh, 2001;

168

Gender and Politeness

Shaw, 2002). Thus, a direct, assertive style of speech, commonly associated with masculinity, is still largely viewed as the appropriate language style for the public sphere. Therefore, assertiveness training programmes developed for women in the 1980s and 1990s often focused on changing women’s language styles, so that, instead of displaying what was characterised as deference and indecision, the woman speaker projected an image of herself, through her language, of being confident and assertive. There are obvious problems with the type of language which is advocated for women in some of these programmes, since they often rely on a number of systematised routines, as Cameron has noted, and furthermore the positive evaluation of direct language is assumed to be unproblematic (Cameron, 1995). However, as a strategic intervention which enabled women in the public and the private spheres to assert themselves linguistically, it is clear that such training had a particular value. Despite the value of this early focus on women and men as different speech communities, as Bing and Bergvall remark It would be ironic if feminists interested in language and gender inadvertently reinforced gender polarisation and the myths of essential female–male difference. By accepting a biological female–male dichotomy, and by emphasising language which reflects the two categories, linguists may be reinforcing biological essentialism, even if they emphasise that language, like gender, is learned behaviour. (Bing and Bergvall, 1996: 18)

Cameron (1998b) argues that the focus on difference-versusdominance approaches to the analysis of gender and language, with the dominance analysts being criticised for problems with their analytical procedures and difference theorists being critiqued for their political shortcomings, leads to a lack of real debate, since theorists have tended simply to set up camps and defend their own positions. She suggests, instead, that dominance theorists should develop more thorough analytical procedures and focus, not on a simplistic notion of dominance as such, but on conflict. Thus, rather than assuming that breakdown in communication between males and females occurs because participants do not understand the intentions of the other speaker, as Tannen does, Cameron argues that we should assume that perhaps it is not misunderstanding which is at issue, but conflicts of interest between certain men and women: conflict over increasingly diminishing resources and power; or conflict over

Theorising gender

169

perceptions of the position from which the speaker is/or should be speaking. Within language and gender research, there has been a wealth of research working within either the difference or dominance frame of reference which has aimed to demonstrate empirically that women or men use a particular feature. The one striking overall assessment which can be made of nearly all of the research done on language and gender differences is that the research is contradictory. The hypotheses are generally very clear: usually taking the format ‘in what way does women’s use of such and such an element differ from men’s use of the same element, when other variables are kept constant’. However, whichever research article seems to prove that women’s language use does differ from men’s language use (for example, that women interrupt less, that they are interrupted by men more, that they use tag-questions more, or directness less), another research project claims that, in fact, in other contexts, men use the same element to the same extent or more than women. For example, in relation to interruption, Chan contrasts Zimmerman and West’s (1975) study, which suggests that men interrupt women more, with Smith-Lovin and Brody’s (1989) study, which suggests a slightly more complex situation where men interrupt the most but women interrupt men just as much as they interrupt other women. Some research seems to find that there are no differences whatever which can be wholly attributed to sex difference alone (Chan, 1992). This is not to say that empirical research should be completely discarded, but it does suggest that factors other than gender may be playing a role in the way that people behave linguistically. It also suggests that gender needs to be approached and conceptualised differently. Language and gender research must move beyond the binary oppositions of male and female. Beyond binary thinking In recent years, gender has begun to be theorised in more productive ways, moving away from a reliance on binary oppositions and global statements about the behaviour of all men and all women, to more nuanced and mitigated statements about certain groups of women or men in particular circumstances, who reaffirm, negotiate with, and challenge the parameters of permissible or socially sanctioned

170

Gender and Politeness

behaviour (Coates and Cameron, 1988; Johnson and Meinhof, 1997; Bergvall, Bing, and Freed, 1996).6 Rather than seeing gender as a possession or set of behaviours which is imposed upon the individual by society, as many essentialist theorists have done so far (see Butler, 1990; Fuss, 1989 for a critical overview), many feminists have now moved to a position where they view gender as something which is enacted or performed, and thus as a potential site of struggle over perceived restrictions in roles (Crawford, 1995). Cameron and Coates’ edited collection Women in their Speech Communities (1988) was one of the first attempts to analyse the specificity of the production of speech by particular groups of women, in particular communities, at specific locations and times. In this collection of essays, the language of older, white women in a Welsh mining community was analysed, as well as the language of British Black women in Dudley. Other researchers started to turn to this type of ‘punctual’ analysis – one which is focused on a specific linguistic community at a particular moment – because the generalisations which had characterised feminist analyses of language in the past were considered untenable. The essays in the collection edited by Johnson and Meinhof on masculinity and language also signalled a change in the focus of language and gender research to analyse women and men’s speech production in relation to each other, rather than in isolation (Johnson and Meinhof, 1997). Bing and Bergvall’s essay ‘The question of questions: beyond binary thinking’, and the collection which they edited, Rethinking Language and Gender Research (1996), is an important move forward in language and gender research, because it seems to draw together a number of discontents which had been surfacing in the research literature over a period of time, and which centred precisely on the difficulty of making generalisations about women as a homogeneous group. They call for a questioning of the clear-cut divisions that researchers had made between the linguistic behaviour of males and females, arguing that the boundaries between women’s and men’s language are not clearly demarcated. They draw an analogy with racial categorisation, where they argue that, particularly in American society, it has become possible to acknowledge the diversity within ‘racial’ groups, and it has become difficult and politically problematic to assert that there is a biological basis for the category ‘race’ at all. In relation to gender, they draw attention

Theorising gender

171

to the variety of sexual identifications which cross the binary divide between female and male, such as hermaphrodite, transsexual, transgendered individual, androgyne, and they thus assert: ‘the simple belief in “only two” is not an experiential given but a normative social construction’ (Bing and Bergvall, 1996: 2). This focus on individuals who do not fit neatly into the male/female category is productive, since it forces us to see that the boundaries of the categories of sex and gender are blurred. We need to extend this thinking about sex and gender to recognise that these categories are also not internally homogeneous. In relation to the previous research in language and gender difference, they feel it necessary to ask the following questions: ‘1. Why are the questions that strengthen the female–male dichotomy so frequently asked, while those that explore other types of variation evoke much less interest? 2. How much of this apparent dichotomy is imposed by the questions themselves?’ (Bing and Bergvall, 1996: 3). Their argument centres on the dilemma within linguistic research that if you analyse data asking the question ‘In what ways do men and women speak differently?’, then that is what you will find. Similarities between the linguistic behaviour of women and men will be ignored or minimised in the interpretation of the results, and so will differences among women. Bing and Bergvall note that few features, if any, can be said exclusively to index gender. They also criticise the way that statistical averages are used to generalise about women’s linguistic behaviour; they argue ‘one obvious oversimplification is that of using statistical differences between two groups as proof that all members of one group have characteristics shared by no members of the other group (and vice versa)’ (Bing and Bergvall, 1996: 15). The problem, for them, is not with difference as such, but with gender polarisation, as Bem argues: ‘it is thus not simply that women and men are seen to be different but that this male–female difference is superimposed on so many aspects of the social world that a cultural connection is thereby forged between sex and virtually every other aspect of human experience, including modes of dress and social roles and even ways of expressing emotion and experiencing sexual desire’ (Bem, 1993: 2, cited in Bing and Bergvall, 1996: 16). Research in gay and lesbian speech styles and in Queer theory has made gender and language researchers question the seeming stability

172

Gender and Politeness

of the term ‘woman’ and ‘women’. Although early research in this area seemed to be trying to assert the difference of lesbian from straight women, more recent work has questioned the assumption that the linguistic behaviour of lesbians can be generalised about (Wittig, 1992). In the essays in the collection Queerly Phrased, edited by Livia and Hall, the very notion of a lesbian language is at once posited and held under erasure, that is, in Derridean terms, its status is called into question (Livia and Hall, 1997). Furthermore, in collections of essays, such as Leap’s Beyond the Lavender Lexicon and at conferences such as the International Language and Gender Association conference at Berkeley, in 2000, the notion of a gay language or a lesbian language has been subject to careful scrutiny, distinguishing stereotype from actual language production (Leap, 1995). Queen, in her work on lesbian speech styles, has found a way of dealing with this seeming paradox (Queen, 1997). She demonstrates that certain lesbians may use the stereotypes of straight feminine speech, as she puts it: ‘there is . . . a stereotype concerning the ways in which women speak, and it is this stereotype that women either aspire to or reject (and sometimes both simultaneously)’ (Queen, 1997: 239). She suggests that by using stereotypes of straight feminine and masculine speech parodically together with stereotypes of the way that lesbians speak, either parodically or in an affirmative way, ‘lesbians create an indexical relationship between language use and lesbian “identity”’ (Queen, 1997: 239). One of the elements which Queen suggests that lesbian women use in this verbal parodic play is over-polite forms, euphemism, and the general avoidance of ‘off-colour and indelicate expressions’ (Queen, 1997: 240). By this juxtaposing and recontextualising of stereotypical straight feminine forms alongside stereotypically masculinist norms such as swearing and directness, lesbians may be creating a distinctly lesbian style of speech.7 That interrogation of the affirmation of the existence of a set of linguistic signs which could be interpreted as signalling to others that one is gay or lesbian, and also of the existence of a set of linguistic patterns used uniformly by gay and lesbian people, has led also to a questioning within language and gender research as a whole that a similar difficulty arises within the analysis of women’s language and men’s language in general, and heterosexual women’s and men’s language specifically. Furthermore, Black feminist linguists have

Theorising gender

173

analysed the elements of Black women’s speech which could be seen to distinguish them from white women’s speech, but these seem to be very context-specific and cannot be said to constitute a form of language which all Black women use in all situations (Morgan, 1999; Henley, 1995). Many feminist linguists and theorists have been influenced by Judith Butler’s work on gender and performativity (Butler, 1993; see also Salih, 2002). She argues that gender is a repeated performance of a range of behaviours associated with a particular sex: ‘The materiality of sex is constructed through a ritualised repetition of norms’ (Butler, 1993: x). Thus, gender is not a given, a possession, but rather a process which one constantly has to perform. Crawford suggests that rather than seeing gender as a noun, we should see it as a verb (Crawford, 1995). The stress on performativity does not suggest for Butler that one can be anything that one decides to be: ‘. . . if I were to argue that genders are performative, that could mean that I thought that one woke in the morning, perused the closet . . . donned that gender for the day and then restored the garment to its place at night. Such a wilful and instrumental subject, one who decides on its gender is clearly not its gender from the start and fails to realise that its existence is already decided by gender’ (Butler, 1993: x). Gender for Butler, therefore, pre-exists the individual and is negotiated with by them in their performance of their identities. This performativity is a constant process rather than something which is achieved: ‘that this reiteration is necessary is a sign that materialisation is never quite complete, that bodies never quite comply with the norms by which their materialisation is impelled’ (Butler, 1993: 2). Thus, although the individual is not in control of the production of its gender identity, there is the possibility of some measure of resistance and hence change.8 This important questioning of the notion of gender does not mean that the category of gender is empty and that there is no such thing as gender difference. For, as Freed (1996) argues, despite the fact that the category ‘woman’ is not one which is coherent, that does not prevent people classifying you as a woman and making judgements about you on the basis of that classification (Freed, 1996). What has to be reconsidered is the simple binary division between female and male, and also the way that gender operates at the level of a system which has been institutionalised, and which operates in

174

Gender and Politeness

stereotypes and assumptions about context which have a material impact on groups as well as individuals, rather than as something which functions simply at the level of the individual. Gender in relation to other variables As I mentioned earlier, one of the problems with early feminist research was that it often focused exclusively on the language usage of white, middle-class women and then made generalisations about all women. Many studies have since shown that groups of women behave in different ways depending on variables of context, class, race, affiliation, familiarity, education, and so on. In some ways, it could be argued that gender itself in isolation does not exist, but only gender as it is inflected by race and class in the way that these elements are worked out in interaction by individuals (McClintock, 1995). Gender is also both the way one experiences oneself as a gendered being, as well as (and in conjunction with) the way that one is treated as a gendered individual. Perhaps one of the most difficult variables to analyse for feminist linguists has been power. As I stated above, implicit in much feminist thinking about gender difference has been a particular model of power relations, which presupposed that there was a more or less simple correlation between males and power and females and powerlessness (Lakoff, 1975; Spender, 1980). Whilst Foucault’s formulation of power relations has been influential in this area, and many feminists have urged that we need to think through power relations in a more complex manner to avoid such simple binary oppositions, there has been until recently little work which details how to analyse seemingly endemic structural inequalities and at the same time individual transgressions and contestations of those inequalities (Foucault, 1978; Thornborrow, 2002). If we consider Foucault’s notion of the dispersion of power (that is, the spread of power throughout a society, rather than the holding and withholding of power by individuals), we might be able to move towards an analysis which sees language as an arena where power may be appropriated, rather than societal roles being clearly mapped out for participants before an interaction takes place (Mills, 1997). In engaging in interaction, we are at the same time mapping out for ourselves a position in relation to the power relations within communities of practice and

Theorising gender

175

within the society as a whole. This is what I call interactional power, to differentiate it from those roles which may or may not be delineated for us by our relation to institutions, by our class position, and so on (Mills, 2002). It is possible for someone who has been allocated a fairly powerless position institutionally to accrue to themselves, however temporarily, a great deal of interactional power by their verbal dexterity, their confidence, their linguistic directness (those more stereotypically masculine/competitive/report talk attributes), as well as through the use of the seemingly more feminine linguistic displays of care, concern, and sympathy, described as co-operative strategies or rapport talk (Coates, 1988; Tannen, 1991).9 This interactional power is a set of resources which is available to all participants regardless of their institutional rank, and which they decide whether to draw on or not in particular contexts. For example, a female secretary in a university department may be able to use a fairly direct form of address to those in positions of power over her, because of her access to information upon which they depend; conversely, lecturers who need this information and who are reliant on her, will need to employ politeness strategies which might normally signal deference (Mills, 1996).10 In constructing a gender identity for oneself, there is also the possibility of choosing simply not to abide by the ‘rules’ which seem to be associated with the context. For example, at a British university where I studied, in the senior common room there was an older white, working-class female server at the coffee bar who swore all of the time, talked extremely loudly on seemingly inappropriate personal subjects, engaged everyone in very lengthy conversations, and made fun of all of the professors. Whilst most people seemed to find her linguistic behaviour extremely trying, they nevertheless tolerated it. She had a fairly ‘low’ position in the university hierarchy, and yet she simply flouted all of the linguistic ‘rules’ of interaction of how one should behave with senior staff. Thus, positions of power mapped out by one’s role in an institution may not relate directly to the interactional power that one may gain through one’s access to information, one’s verbal skill, or one’s display of care and concern for other group members. Consider the following example of an exchange between a white, middle-aged female educational adviser, E, and her white, middleaged female secretary, D, working in a County Council education office. Whilst there is a clear distinction between them in terms of

176

Gender and Politeness

institutional status, they have known each other and worked together for many years. D is describing a form which she has drafted which needs to be sent out in E’s name to schools for information about reading age: 1D: 2E: 3D: 4E: 5D: 6D: 7D: 8D: 9D: 10D: 11D: 12E: 13D: 14D: 15D: 16D: 17E: 18D: 19D: 20E: 21D: 22E: 23D 24D: 25D: 26D: 27D: 28E: 29D: 30D: 31E: 32D: 33E: 34E: 35E: 36D: 37D: 38D:



40E: 41D:

he was suggesting that perhaps I should be asking for more (.) and then [Hmmm] select from that= =Hmmhmm= =um I don’t know whether I could get more information than I’m asking for(.)so this is the kind of thing that could come um with um an explanatory letter um I’m also wondering whether we might be able to find some controls as well I’ll explain about that later(.) and I thought I could design it so that um if schools felt a bit uncomfortable about er confidentiality um what they could have(.) the information at the top that we’re interested in at the top and just in order to identify them [mmm] could fill in this our information would be there for them on files and that’s(.) that’s all the information we need and if they identify the reading information they could then cut off the other information and then just send that back to us= =oh I see yes and then= =because we’re not interested in individual children it’s just the(.)you know how many children with language problems how many children [I’m with you] with this have a severe reading problem for example or whatever= =yes this is for your(.) your filing system? [so] that’s all the information we need for our files um because that would tell us about. the kind of problems (.)yes I’ve left that back because I wasn’t (.) this again I’m I wasn’t sure what kind of information [ well one must always go into chronological [ your schools will translate it into [ chronological age reading age= =reading age uhuh= =standard deviation of any one test is really it’s the same thing for every test isn’t it? that’s the one that’s always quoted in the manuals zero point three or whatever [um] that’s right it’s to do with the percentages of children falling within certain [ ah that’s a percentile= =um(.) yes it relates to that [

Theorising gender

177

42E: 43D: 44E: 45E: 46E:

standard deviation(.)we’ve never used it though= =uhuh= → =if you ask a class teacher about in context of a test a standard deviation it’s the degree of error of the test zero point three on a Young’s is a standard deviation(.) you’re talking about a deviation from [ 47D: the 48D: deviation from the mean [ 49E: from the main [ 50D: a hundred would be the mean and one standard 51D: deviation usually [ → 52E: we never use it that way [ 53D: usually is er= → 54E =ah well it’s never used that way in any of our tests so you’d be better 55E: off= 56D: =well if we get down to 85 that is minus one standard deviation usually 57D: I mean 150 plus I= → 58E: =mm are you sure you’re right? not between 90 and 110? 59D: well well it it 60D: would depend yes it would depend it depends on the scale you’re using 61E: [anyway] → 62E: yeh that’s as an aside but on ours er er you’re going to throw most 63E: teachers with this they’re going to want chronological age 64D: [uhuh] 65E: reading age then if you use a column with difference between in 66D [hmmhmm] 67E: months= → 68D: =I really need some kind of standardised score= 69E: =well the rest of the schools from the whole of the year band or 70E: whatever (.)you’ve got two children who are beneath their 71D: [mm] 72E: chronological age= 73D: =mmm= 74E: =er very young there so you’d expect= 75D: =so say for example you put [ → 76E: what which class was your son in last 77E: year?= → 78D: =er er what I would need to know is er you know if they’re sort of four 79D: months behind how significant is that considered er a problem is that 80D: considered within the normal range um= 81E: =it’s changing all the time you see 82D: yes Data: 54.3/4 EXAMPLE 10

What we expect in an interaction between a female manager and her secretary if we focus on their institutional power relation, according

178

Gender and Politeness

to Brown and Levinson’s model, is that the female secretary should display deference in her speech which here she signally does not do. This is a very pacey conversation, with a great deal of quickly switching turns. D provides a great deal of the information in this interaction, and although she is slightly more hesitant than E this on its own cannot be said to signal deference or a recognition that she is in an inferior position in the hierarchy. Hesitation cannot be said to have only one function, as many conversation analysts have assumed. D asserts her point of view about the way that the form should be composed and the information in it, for example: →

67E: 68D: 69E:

months= =I really need some kind of standardised score= =well the rest of the schools from the whole of the year band or

She interrupts, corrects, and disagrees with E consistently throughout the interaction. Occasionally, she uses a mediated disagreement for example: →

40E: 41D:

ah that’s a percentile= =um(.) yes it relates to that

when E asserts that something is a percentile and D does not say that this is not so, but rather suggests there is a relation. However, she completely ignores an interruptive direct question from E: 75D: → →

76E: 77E: 78D:

=so say for example you put [ what which class was your son in last year?= =er er what I would need to know is er you know if they’re sort of four

E asks her a personal question about her son, but D ignores this at line 78 and simply continues to talk about the way the form is to be set out. E is also fairly assertive and states directly that the type of information D is requesting will not be effective, stating ‘we never use it that way’ as a reason why D’s suggestion should not be adopted. → →

52E: 53D: 54E 55E: 56D:

we never use it that way [ usually is er= =ah well it’s never used that way in any of our tests so you’d be better off= =well if we get down to 85 that is minus one standard deviation usually

Theorising gender →

57D: 58E:

179

I mean 150 plus I= =mm are you sure you’re right? not between 90 and 110?

And in line 58, when D does not immediately submit to her suggestion, she then suggests that D has made a mistake in the extent to standard deviation. And in line 62, when D persists, E suggests that it is not only E who disagrees with her, but all of the teachers using the form: →

62E: 63E:

yeh that’s as an aside but on ours er er you’re going to throw most teachers with this they’re going to want chronological age

Thus, this would seem to be more of a conversation between equals than between a director of education and her secretary. For example, they seem to overlap a great deal in this discussion, and Coates argues that this completion of others’ turns may signify a degree of equality in a relationship between women (Coates, 1988). If we follow Tannen’s and Holmes’ view of the way that women interact, we should expect both participants to use co-operative strategies, mitigating the force of their assertions and trying not to disagree openly (Tannen, 1991; Holmes, 1995). However, in this example, both interactants follow the norms of professional business interaction in that they confront each other directly and forcefully about what type of information should be included, but at the same time, because they have known each other and worked together for a number of years, this confrontational style is not considered impolite, and they also at other points in this conversation draw on more co-operative styles of speech in the process of working out which information is needed on the form. O’Barr and Atkins, in their 1980 paper ‘“Women’s language” or “powerless language”?’, argue that there is a confusion between the language features which are determined by gender and those determined by a position of lesser power.11 They argue that in their analysis of the type of speech that is produced by female and male witnesses in a courtroom setting, powerless men seem to produce speech which exhibits the same features that, it was argued by Lakoff, women in general use (Lakoff, 1975). They also show that not all women use the features that Lakoff stated were indicative of women’s language to the same degree. Thus, they argue that ‘socalled “women’s language” is neither characteristic of all women nor limited only to women’ (O’Barr and Atkins, 1980: 102). All of the

180

Gender and Politeness

women in the courtroom who used relatively few of the ‘feminine’ ‘powerless’ features described by Lakoff were of high status, primarily middle class and professional. The men who used high numbers of these features were low status, mainly working class, and unused to courtroom protocol. O’Barr and Atkins suggest that power relations play a more important role than gender, as such, in the production of certain types of language; however, even this statement must be treated with some caution. It is clear that power, however we define it, is part of the way that gender as a whole is defined and constituted, therefore power cannot be entirely disentangled from gender. Nor can power be considered in isolation from other variables such as race, gender, and class. As O’Barr and Atkins remark: ‘It could well be that to speak like the powerless is not only typical of women because of the all-too-frequent powerless social position of many American women, but is also part of the cultural meaning of speaking “like a woman”. Gender meanings draw on other social meanings’ (O’Barr and Atkins, 1980: 111). Thus, what we need to move away from is the sense that all women are powerless and all men are powerful, and we also need to question the way that we define power. Ariel and Giora (1992) examine this notion that power might be a more important determinant of ‘deferent’ behaviour than gender alone, but they suggest that the distinction between in-group and out-group is key to whether deferent linguistic behaviour is adopted. This enables us to see some flexibility in relation to gender as there may be times when some women consider certain other women to be within their in-group, and, at other times, they might see their allegiances as being with particular males. However, whilst Ariel and Giora seem clear that power and in-group/out-group relations are correlated, the relation between them is far more complex. It cannot be assumed that all in-group members will be of equal status or that they will always co-operate, nor can it be assumed that all out-group members will be treated antagonistically on all occasions and to the same degree. Cameron (1998a) argues for a more useful approach to the analysis of power and gender, focusing less on unchanging, unequal relations between men and women but rather on the resources available to speakers in particular positions to draw upon strategically. Her approach ‘treats the structural fact of gender hierarchy not as

Theorising gender

181

something that must inevitably show up in surface features of discourse, but as something that participants in any particular conversation may, or may not, treat as relevant to the interpretation of utterances. Furthermore, it insists that where assumptions about gender and power are relevant, they take a form that is contextspecific and connected to local forms of social relations: however well founded they may be in structural political terms, global assumptions of male dominance and female subordination are too vague to generate specific inferences in particular contexts, and thus insufficient for the purposes of discourse analysis’ (Cameron, 1998a: 452). However, whilst it is important not to over-generalise about men and women, when we focus on the particular ways that men and woman interact, we must nevertheless see that those structural inequalities, and the stereotypes that we hypothesise on the basis of our knowledge of these inequalities, do play a role in the way that the interaction takes shape. Where many studies falter in the analysis of the relation of power and gender is in the assumption that there is a simple relation between them. Although it may be possible to make generalisations about the types of language which will be produced when there are differences of power and status, there is clearly no simple link between, for example, interruption and power difference (Thornborrow, 2002). Many theorists assume that because males generally have more power within a particular community that they will necessarily interrupt more, because interruption is the prerogative of the powerful. Whilst there is a stereotypical element of truth in this assertion, the results of the research are far more complex and often go against this hypothesis, both that there is a clear correlation between power and interruption and, following on from that, that males interrupt females more frequently (Chan, 1992). However, if males in a particular interaction do interrupt the female interactants more, then that may well be significant in gender terms within that interaction.12 As I mentioned in chapter 3, there has been little consideration of the importance of race and class as factors which interact and intersect with gender, because of the focus in mainstream Western feminist theory until relatively recently on the behaviour of white, middle-class women’s values and needs; this has resulted in the relegation of other groups of women to the status of minority groups

182

Gender and Politeness

(Lewis and Mills, forthcoming; Sandoval, 1991; Mohanty, 1984; Minh-ha, 1989; Frankenberg,1993; Afshar and Maynard, 1994). Within feminist theory in general there has been a shift in focus, from assuming that race necessarily makes a difference to language production and interpretation, to working with the idea that race can, but may not, make a difference. Race and ethnicity are no longer discussed in the rather ‘blanket’ way that they were in the past: firstly, because there are innumerable differences in the cultural groups which are labelled as racially similar; and, secondly, because of the challenging of the stability of the term ‘race’ itself, since this seems to be embedded in a nineteenth-century ‘scientific’ racist ideology (Jarrett-Macaulay, 1996; Safia Mirza, 1997). What feminist theory is now focused on is the specificity (and perhaps also the instability) of difference, and challenging the notion that perceived difference will necessarily be the result of racial difference alone. Thus, analysis of the language of particular groups of women in particular locations is undertaken rather than analyses of black women or white women as homogeneous groups (Nichols, 1998; Bucholtz, 1996; Edwards, 1988). In this way, we can analyse the intersection of factors such as social mobility and employment patterns with factors such as race, as Nichols has in her analysis of the very heterogeneous speech patterns of Black women in coastal South Carolina, where she found elements of both conservative and innovative usage (Nichols, 1998). Furthermore, feminist theory is now focused as much on the specificity of heterosexuality, middleclassness, and whiteness, and the privileges which go along with these positions, as it is on the specificity of blackness, workingclassness, and lesbianism, so that the term ‘race’ is no longer used solely to refer to the analysis of Black women, but is used to analyse women in relation to racial (and other) differences (Frankenberg, 1993; Wilkinson and Kitzinger, 1993; Maynard and Purvis, 1995). Furthermore, alongside studies which analyse the privileges of white, middle-class women and the problematic generalisations made from their linguistic behaviour to women as a whole, there has developed a concern, particularly important in the context of this study, also to analyse the way that the speech patterns stereotypically associated with white, middle-class women’s speech are very often stigmatised and subject to mockery. As I mentioned in chapter 3, class is an important variable which has to be considered in relation to gender. However, it is not an easy

Theorising gender

183

variable to analyse, since even now it is difficult to assign women to a class position easily.13 Furthermore, women’s relation to their assigned class position varies greatly, for example some working-class women choose to aspire to middle-class values and reject workingclass values and culture, whilst others reject and ridicule middle-class values. Discussions of class are made further complicated because there seems to be a tendency to assume that middle class equals feminine and working class equals masculine, thus making analysis of speech difficult without drawing on, or being influenced by, these stereotypes.14 As I mentioned above, many of the analyses of women’s language have focused on middle-class norms, and the speech of groups of working-class women has only come to the fore in recent years, through the analysis of dialect and accent.15 The generalising of stereotypes of middle-class, white women’s speech to other groups has been challenged (Coates and Cameron, 1988). Indeed, the specificity of middle-class, white women’s experience has begun to be analysed more in recent years (Davidoff and Hall, 1995). Skeggs, in her 1997 analysis of British working-class, white women, has argued that assessment by others (real and imagined) is a very important factor in the constitution of their class positions. The assessment as ‘common’ and ‘respectable’ is one which weighs on a wide range of behaviours from child-rearing, drinking, sexual behaviour, physical appearance and weight, and linguistic behaviour, such as swearing and loudness. Assessments of respectable behaviour have the effect of separating the person from the ‘rough’ working class and moving them into a position of alignment with the lower middle class, whereas assessments as ‘common’ align the person with lower working-class positions, which for women are intensely problematic, especially in relation to particular types of femininity and sexuality. Issues of language are crucial in the daily representation of oneself as classed, raced, and gendered: Skeggs argues that ‘to be working-classed . . . generates a constant fear of never having “got it right”’ (Skeggs, 1997: 6). To be judged to be a working-class woman by middle-class people is always not simply to be classified into a class position but also to be categorised as inadequate and inferior: ‘in relation to language and this sense of “proper”, “lady-like” usage, this is of importance, as working class women may feel that in order to aspire to respectability, they need to use particular types of language, formalised politeness being one of the strategies available . . . However, other working class women will

184

Gender and Politeness

refuse to appropriate these middle class norms to make them their own – not being middle class is certainly valued in many working class social groups’ (Skeggs, 1997: 11). Thus, the use of certain forms of language, such as directness, loudness, and swearing, associated by the middle classes with impoliteness, may well be part of a strategy to mark oneself off from those middle-class norms of feminine behaviour and to affiliate oneself with working-class values. Therefore, rather than considering working-class women to be a homogeneous group, we should see that constructing one’s gendered, raced, and classed identity involves taking a stand on language. This may indeed not be a consistent stand in one’s own language repertoire. Gendered stereotypes: femininity and masculinity Drawing on Bourdieu’s notion of habitus, we can define a stereotype less as a fixed set of characteristics than as a range of possible scripts or scenarios, (sets of features, roles, and possible narrative sequences), that we hypothesise (Bourdieu, 1991). Thus, we hypothesise some extreme aspect of an out-group’s perceived behaviour and generalise that feature to the group as a whole. In this sense, the stereotype is based on a feature or set of behaviours which may have occurred within that community, but the stereotype is one noticeable form of behaviour which is afforded prototypical status, backgrounding all of the other more common, and in a sense the more defining, forms of behaviour (G. Lakoff, 1987; see Mills, 1995a for a discussion of scripts and scenarios). This notion of the prototype is quite important, since hypothesisation of stereotypes often informs judgements made about male and female linguistic behaviour and sets for us, often unconscious, notions of what is appropriate. The notion of the prototype allows us to acknowledge that stereotypes of femininity which circulate within British society now may have originally been descriptions of certain aspects of white, middle-class women’s behaviour within a certain era, but that even with that class, at that time, there were other forms of behaviour which conflicted with and challenged them. This stereotype is not a fixed set of behaviours which exist somewhere, but the hypothesised version of the stereotype is something which is played with by those arenas where our ‘common’ experience is mediated, for example on television, in advertising, newspapers, and magazines. It is clear that

Theorising gender

185

we as a nation do not share experience, but the media work on the assumption that we can consider certain types of information as ‘common’ to all readers/viewers. Members of audiences, however, take up a variety of positions in relation to this information, some affiliating with the values of the stereotype and others rejecting them. The hypothesised forms of stereotypes are equally damaging to both males and females, since they consist of assumptions about us which often clash with our own perceptions of ourselves. These stereotypes are often authorised in some sense through being mediated by the media and thus they have an impact on us; they are not simply someone else’s personal opinion of us but they are also backed up by institutions. Thus, the stereotype that women should take the major role in childrearing and household management is one which is challenged by many oppositional discourses such as feminism; nevertheless, it is still a stereotype which can be activated by many men when considering their own roles within the household, because it is still kept active by certain groups within the society and implicitly authorised. These stereotypes of appropriate behaviour for males and females have been challenged by feminism, so that the notion that women are weaker than men or that they should not compete with men in the workplace are notions which are under challenge.16 To give an example of stereotypical assumptions, let us consider the analysis of an anecdote by Cameron (1998a). A friend’s father, when he sits down to eat his dinner, always asks his wife: ‘Is there any ketchup Vera?’ and this indirect question is interpreted by all as a request by the man for his wife, Vera, to fetch him the ketchup. Conservative stereotypes of the role of wives in relation to husbands, which here are shared by both the wife and the husband, lead to both interpreting this as a request for the ketchup to be brought rather than as a request for information about the availability of ketchup. However, their feminist daughter is angered by the way in which the couple collude in this stereotypical behaviour and she draws attention to the way in which this type of requesting behaviour only works if we assume that women’s role is to serve men. Hypothesisation of stereotypes of gender, as I have shown above, is very powerful in our assessment of language both as interactants and as analysts. However, that is not to say that there is only one

186

Gender and Politeness

stereotype of women’s linguistic behaviour or that stereotypes are fixed. If we consider the stereotypes of the nagging woman and the gossip, these can be seen to coexist with other stereotypes of women which are not concerned with excessive linguistic production and excessive demands, for example, the stereotype of the overpolite woman who is concerned only with surface appearances, or the stereotype of the self-effacing woman silenced by a dominating male partner. As Liladhar has shown in her work on femininity, feminist analyses of femininity have changed markedly over the last ten years, where femininity is no longer seen as a set of negative behaviours which kept women in a subordinate position, but feminists are beginning to see the potential play within the behaviours which have been traditionally seen as denoting powerlessness, particularly when they are used ironically as in the demeanour of the Soap Queen and the Drag Queen (Liladhar, 2001). Whilst in the past femininity seemed to denote a concern with one’s appearance to the detriment of one’s intellect, femininity now seems to denote a range of stereotyped behaviours which can be ironised and played with (Bell et al., 1994). Skeggs argues that ‘femininity brings with it little social, political and economic worth’ (Skeggs, 1997: 10). In that feminine behaviour is not generally valued, we might be led to ask why women do in fact orient themselves to such behaviour, as there are women who are more feminine-affiliated than others (Crawford and Chaffin, 1986; Gilbert and Gubar, 1988). However, Skeggs has shown that, for example, in relation to caring, which is an important aspect of femininity, it is one way of achieving some sense of value when in a position of relative powerlessness ‘a caring identity is based not only on the fulfilment of the needs of others and selflessness but also on the fulfilment of [working-class women’s] own desire to feel valuable’ (Skeggs, 1997: 62). Thus, even though the adoption of feminine positions does not offer great value within the society as a whole (caring jobs are not economically rewarding), they may, however, define working-class women in ways which are of value, for example, they may construct a woman as respectable and therefore aligned to what are seen as middle-class values. Through the alignment with middle-class values through what is seen as femininity, many middleclass women can gain some power and assert their difference from other groups of what are for them ‘non-feminine’ women. Thus,

Theorising gender

187

investment in femininity provides some status and moral position, both in relation to working-class and middle-class women. Furthermore, in previous eras, conventional femininity, whilst not exactly valued by the society as a whole, was at least expected as a behavioural norm. Now, however, it seems as if the representation of stereotypically feminine women is rarely presented on radio or TV without mockery or ridicule. These stereotypes of gender are important in the process whereby we assess our Others. Cameron asserts ‘Information about who someone is and what position she or he speaks from is relevant to the assessment of probable intentions. Since gender is a highly salient social category, it is reasonable to assume that participants in conversation both can and sometimes (perhaps often) do make assumptions in relation to it’ (Cameron, 1998a: 445). But, as Cameron makes clear in her work, whilst we may be making assumptions about gender in our interactions, stereotypes of gender, because hypothesised rather than actual, may not be shared. Where conflict in conversation often occurs is when assumptions about gender are not shared by participants, and this is not a conflict which is restricted to a struggle between women and men, as the dominance theorists assumed, but can be a conflict between women, where some hold a more traditional view of what women should do and how women should speak, whilst others aim to challenge those stereotypes. Femininity has often been associated with the private sphere and the values associated with that sphere. Therefore, caring, concern for appearances, emotional excess, incompetence in relation to nondomestic tasks have all in the past been markers of the feminine; however, with the changes which have taken place in the last twentyfive years in relation to women’s employment within the public sphere, these elements of femininity are more difficult to maintain. Greater social mobility, greater choice in relation to marriage, divorce, and conception have made major impacts on women, and whilst many women would still not openly identify as feminist, nevertheless many of the values of feminism have infiltrated common sense. That does not mean that the ideals of femininity have simply disappeared because they are constantly invoked, sometimes ironically, but often in contradictory ways in relation to this commonsense feminist ideology. Halberstam’s (1998) work on female masculinity has been important in mapping out forms of behaviour and

188

Gender and Politeness

style available to women other than conventional feminine forms, and Holland (2002) has shown that women can appropriate notions of femininity to describe their own forms of dress and behaviour which seem directly to challenge feminine values. Thus, one of the many important advances made by feminism is to open up within the notion of what it means to be a woman a distinction between femininity and femaleness, so that one can be a woman without necessarily considering oneself to be (or others considering one to be) feminine. Masculinity has often been posited as the direct opposite of femininity. One of the defining features of masculinity is seen to be aggression, which is often considered to be a biological part of being male (caused by testosterone), rather than as a set of characteristics which are acquired in a complex negotiation between the individual and what they hypothesise to be the values of their communities of practice and the wider society. Masculinity is often described in terms of battle and warfare. Stereotypically masculine speech is seen to be direct and forceful, arguments between males are described as ‘cut and thrust’ or as verbal ‘sparring’ (Coates, 1997; Pilkington, 1998). Tannen’s (1991) work also seems to characterise masculine speech as a speech style aimed at establishing a position in the hierarchy and getting the better of your opponent. De Klerk characterises ‘high-intensity’ masculine language as ‘constituted by dominance, interruption, disputing and being direct’ (De Klerk, 1997: 145). Swearing seems to have a stereotypical association with masculinity, and indeed most of the studies of swearing have concentrated on adolescent males. However, for many men, this characterisation of the ‘hard man’ is not necessarily one which they want to adopt wholesale, but neither do they want to adopt the persona of the ‘new man’. Edley and Wetherall (1997) have described the way that young men ‘exploit the critical or rhetorical opportunities provided by the subject position of the “new man”, not necessarily to claim the position of the “new man” for themselves, but rather to construct a way of being a man which does not involve wholesale adoption of “macho” or “new man” forms of masculinity’ (Edley and Wetherall, 1997: 208). Because of changes in men’s and women’s employment patterns and involvement in the public sphere, together with the impact of feminism, there is a sense in which men, at least at a stereotypical

Theorising gender

189

level, are often represented as in crisis about their masculinity. Whelehan argues that ‘there is much evidence in recent years that men as a group are feeling more disenfranchised by increased unemployment, and the figures for the incidences of violence and suicides among young men are frighteningly high. The popular press speak of the ‘feminising’ of the workplace as one cause of increasing male unemployment, clearly signalling that the more women make up a significant part of the workforce, the more men have to pay . . . men are undergoing a crisis in the way their identity is defined, and this crisis is alleged to be directly related to female emancipation. Feminism is roundly viewed to be at fault. While it is true that the ‘new lads’ are assuredly the product of identity crises, it is not just generated by feminism, but also by gay liberation and anti-racist movements, which act as a reminder of what mainstream male culture, such as big budget competitive sport, regularly excludes’ (Whelehan, 1999: 61). Thus, macho masculinity is considered to be a set of valued or problematic behaviours which are under threat from changes in the behaviour considered appropriate for women and homosexual males. This challenge to masculinity has been embraced by some as a positive opportunity for men to explore different aspects of their identity, but others have seen it as intensely threatening. Thus, media stereotypes such as the ‘new man’ (the feminised and often mocked mythical figure) and the ‘new lad’ (the man who rejects this feminisation and embraces patriarchal values ironically) are available for men and women to react against and incorporate into their own sense of appropriate behaviours. Not all males feel comfortable with masculine speech-styles, for example Stearns comments: ‘Malely male gatherings confuse me a bit; they leave me feeling out of place. Gratuitous obscenities strike me as an unilluminating form of speech and I cannot hold my own in skirt-lifting stories. I have always, in sum, viewed manhood with a bit of perplexity’ (Stearns, cited in De Klerk, 1997: 145). Particularly given the changes that have taken place in terms of the social position of women, many men’s attitudes to women have changed considerably. But these changes have also brought about the rise of ‘laddish’ behaviour and backlash (Whelehan, 1999). Nor should we assume that there are not differences within the types of stereotypes hypothesised for particular groups of males, as Jackson has demonstrated in his analysis of the sexualised representations

190

Gender and Politeness

of black males (Jackson, 1994). Furthermore, many men may feel forced to engage in stereotypical masculine speech behaviour because of the fear of being labelled homosexual by others in their community of practice, as Cameron has shown (Cameron, 1997). We should not assume, therefore, that stereotypes are permanent unchanging discursive structures, but we should see them rather as resources which can change fairly rapidly, with certain anachronistic aspects being available to be called upon by certain speakers. In an article on discursive anachronism, I argue that discursive structures, by their very nature, because they are constantly being challenged and used in new ways by speakers and texts, are in a process of continual change; however, certain of these structures seem as if they are more stable because they have endured over a relatively long period of time (Mills, 1995b). I would argue, however, that it is perhaps the community members’ interactions with these seemingly more stable stereotypes and discursive structures in general which change and thus colour a speaker’s use of them as part of his/her linguistic resources or assumptions. Stereotypes of gender, developed in the interaction between the individual and the society as a whole, and within specific communities of practice, inform individual choice of linguistic style, strategy, and content, either in terms of reaffirming or challenging those stereotypes in relation to one’s own linguistic production or in relation to someone’s assumptions about one’s own gendered identity. Strong women speakers As I have argued so far, for many early feminist linguists, female speech was seen to be powerless speech; however, it is clear that, because of changes in the way that many women perceive themselves and the employment of women in the public sphere, this is no longer the case for all women (Mills, 1998, 1999). There are thus a great many women whose linguistic behaviour does not appear deferent and submissive, and indeed submissive speech by women is now generally derided in the media. Halberstam has analysed women who reject the presentation of a feminine self both in their speech, clothing and their body (Halberstam, 1998) Many linguists have also noted that this powerless speech style is not a norm for African American women (Bucholtz, 1996). Bucholtz describes the

Theorising gender

191

language of African American women in media discourse, analysing the code-switching between Black American Vernacular and standard American English which sets up powerful oppositional positions for African American women (Bucholtz, 1996). Research in Poland and Japan has shown that the language usage of certain groups of women, particularly those who are in paid employment or higher education, has changed dramatically over recent years, especially in relation to politeness and hypercorrection (Baran and Syska, 2000; Okamoto, 1995, 2000). Baran and Syska describe the new speech style adopted by many Polish women as harsh, clipped, matter of fact, impatient, and rude; and young Japanese women are seen to be challenging the deferent, self-effacing forms of speech generally associated with women. Thus, these styles of women’s speech have more in common with stereotypical masculine speech. In the UK, on the radio and television, we frequently hear women ministers, presenters, and spokespersons speaking confidently and competently, using direct forceful language (for example, Madeleine Allbright, Mo Mowlam, Sara Cox). Female singers now do not necessarily use stereotypically feminine language, as Haugen has shown in his analysis of female gangsta rappers, where they overtly position themselves as ‘unladylike divas’, ‘showin’ and provin’ that even a lady can be evil’ (cited in Haugen, 2000). Whelehan argues that Girl Power, associated with groups such as the Spice Girls, ‘seems to involve meeting aggression, particularly sexual aggression, with similar aggression’ (Whelehan, 1999: 45). In an article in FHM, a male magazine, in a discussion of the ladette, the author stated that ‘women who are one of the lads are the men we’re supposed not to be’ (cited in Whelehan, 1999: 49). This aggressive speech behaviour is in direct contrast to the stereotype of polite submissive feminine linguistic behaviour. However, rather than seeing the speech of women in positions of power within the public sphere as simply an appropriation of masculine speech, Webster has shown that the speech of the British ex-Prime Minister Margaret Thatcher mixed features associated with the stereotypical language of both women and men, sometimes within a single utterance. Thus, for a complex range of motivations and judgements made about her audience and her own standing, Thatcher seemed to draw strategically on both masculine and feminine speech elements. Because of the planning that goes into the

192

Gender and Politeness

production of the speech of the Prime Minister in Britain, as Fairclough has shown in his analysis of the speech of Tony Blair, it is fair to assume that the production of Thatcher’s speech cannot be assumed to be typical of other women’s speech (Fairclough, 2000). But this mixing of feminine and masculine styles has been noted of women’s speech in other public sphere domains (Walsh, 2001). Mary Robinson, for example, the EU commissioner, manages to combine forcefulness and a soft-spoken style, as have other women politicians such as Benazir Bhutto. For radio presenters such as Zoe Ball or Sara Cox, on BBC Radio 1, their ‘ladette’ behaviour bears striking similarities to Thatcher’s, in that they both mix elements of stereotypical femininity and masculinity. Cox hams up her Lancashire accent at times, swears, behaves in a ‘sexist’ way towards male callers (for example, calling them ‘gorgeous’), and she talks about excessive drinking, going to night clubs, and watching football. She engages in a great deal of verbal duelling and play, mocking callers to her show. Cox’s performance is little different from that of Chris Moyles, another presenter on Radio 1, who is noted for his stereotypically masculine preoccupations. Cox does, however, also draw on resources which are more stereotypically feminine, for example, describing in detail her relationship with her puppy and apologising excessively when something goes wrong on the show.17 We might consider here the linguistic behaviour of Anne Widdicombe, a British Conservative Shadow Cabinet member, who seems to have developed a particular speech style for herself which is masculine, combative, direct, and forceful. The British press variously describe her and her speech style in both positive and negative terms, either as too forceful (when she takes very right-wing positions, very dogmatically) or sufficiently statesman-like, and indicative of a potential future leader of the Conservative party (when she appears to be articulating what is seen to be ‘common sense’). Since Thatcher, it does seem to be the case that it is slightly more acceptable for women to be forceful verbally in public life and this is in large measure because of the number of women in high profile public positions, in the media, business, and politics, and because those women who are in these positions have adopted, more or less wholesale, the speech styles of the men who still dominate in these public positions. We also need not necessarily see forceful speech as being a positive way of speaking. Adams (1992) remarks on the way that the

Theorising gender

193

discourse norms associated with a particular context, in the case of her research on the broadcast television interview, will play a major role in determining what styles and strategies may be adopted by interactants. Her article questions the notion that an aggressive style in debate (a stereotypically masculine style) is necessarily seen as most effective by participants and observers alike; she notes that those candidates in television interviews who observe the rules in terms of turn-taking, what she terms ‘accruing power by obeying rules’ in short, behaving in a more feminine way, also bring benefits to themselves in terms of how they are judged by the audience (Adams, 1992: 9).18 One way in which women may negotiate a powerful position for themselves linguistically is through the use of swearing. There seems to be a taboo on women swearing which it would appear is changing somewhat. ‘Nice’ women are expected not to swear, and this is a marker of what Skeggs terms respectability, those values that aim to distinguish between ‘rough’ or ‘promiscuous’ working-class behaviour and feminine, middle-class behaviour (Skeggs, 1997). Thus, as I argued above, a concern with respectability and ‘proper’ behaviour is not only a gender issue but one which is inflected through class. However, paradoxically, many women in positions of power use swearing to affirm their position. Madeline Allbright, the American ex-Secretary of State, in a BBC Radio 4 interview in 2001, stated that she nearly gave Colin Powell an aneurysm with her ‘bad language’. And when she saw Cuban fighter pilots on video celebrating after shooting down an American plane she famously said in an interview: ‘That’s not cajones (balls), that’s cowardice.’ Mo Mowlam, the ex-Northern Ireland Minister, was reputed to have used a great deal of ‘bad’ language in meetings. Thus, using seemingly masculine language may be a way for these women of marking themselves out as competent and powerful through showing that they are not restricted by conventional femininity. However, Walsh has noted that those women who have entered into male-dominated professions and have either changed the dominant ethos of the organisation through their ‘feminine’ language use, or who have simply adopted the masculinist norms, have all largely been very negatively viewed by others within that domain (Walsh, 2001). For example, the presenter Ann Robinson’s aggressive, humiliating language behaviour on the British television quiz show ‘The Weakest Link’ has been

194

Gender and Politeness

widely criticised in the British media, whereas aggressive tactics by male presenters such as Jeremy Paxman have not been criticised in such personal and damning terms (Braid, 2001). Thus, the context and community of practice within which speech takes place is crucial in determining the way that speech will be judged. Hypotheses about stereotypes of gender may be a factor in the assessing of the appropriate language for a particular situation, but there is nothing to suggest that it is the only factor which is salient. Thus, if the situation is one in which masculine speech norms have been prevalent over a period of time, it is likely that women who work within the environment will adopt those norms if they are to be seen as professional. Alice Freed suggests in her analysis of the types of speech which are produced by close friends that certain styles of interaction are coded by the participants as feminine; thus, because of the context and the perception that intimate conversation is feminine, the males in her study seemed to be behaving in a stereotypically feminine way (Freed, 1996). This does not seem entirely satisfactory since it is clear that some males would perhaps see this as an occasion to mark their speech in hyper-masculine ways.19 Furthermore, not all linguistic communities would code this type of relaxed conversation as feminine. However, the notion of gendered domains is important here in being able to describe the way that gender impacts at the level of the setting and context, rather than simply at the level of the individuals involved in the interaction. It is clear that context, broadly speaking, is important in terms of the production of speech and in the assessment of what types of language and speech styles are appropriate. Take, as an example, McElhinny’s analysis of the language of women police officers in Pittsburgh, where more masculine practices are adopted, in order for these women to appear professional and credible to the wider community (McElhinny, 1998). She suggests that ‘institutions are . . . often gendered in ways that delimit who can properly participate in them and/or how such participation can take place . . . Workplaces are gendered both by the numerical predominance of one sex within them and by the cultural interpretation of given types of work which, in conjunction with cultural norms and interpretations of gender, dictate who is understood as best suited for different sorts of employment’ (McElhinny, 1998: 309). She argues that because of this gendering of the context, and because of the associations of masculinity

Theorising gender

195

with the type of work which is done, women police officers have to adopt certain types of language style which are associated with masculinity. Because police officers often have to deal with trauma on a daily basis, they generally adopt a particular habitus of affectless behaviour. Women police officers thus have to adopt this style of response to situations which seems to many prototypically masculine, but ‘because masculinity is not referentially (or directly) marked by behaviours and attitudes but is indexically linked to them (in mediated non-exclusive probabilistic ways) female police officers can interpret behaviours that are normatively or frequently understood as masculine . . . as simply “the way we need to act to do our job” in a professional way’ (McElhinny, 1998: 322). Thus, women working within environments where men predominate, and thus masculine norms of behaviour and speech also predominate, are forced in order to be viewed as professional to adopt those norms; however as we have seen within particular environments there is some degree of negotiation with those norms. Feminist linguists and communities of practice As I mentioned in chapter 1, the notion of communities of practice, formulated by Wenger (1998), and developed in relation to language and gender research by Eckert and McConnell-Ginet to particular effect, is useful in the context of the analysis of gender and language (Eckert and McConnell-Ginet, 1998, 1999). Within this view, feminist linguistics should be concerned not only with analysing individual linguistic acts between individual (gendered) speakers but also with the analysis of community-based perspectives on gender and linguistic performance, which in the case of politeness must therefore involve a sense of politeness having different functions and meanings for different groups of people. Eckert and McConnellGinet argue that ‘a community of practice is an aggregate of people who come together around mutual engagement in some common endeavour. Ways of doing, ways of talking, beliefs, values, power relations – in short, practices – emerge in the course of their joint activity around that endeavour’ (Eckert and McConnell-Ginet, 1998: 490). The crucial dimensions of a community of practice are that it will have ‘mutual engagement; a joint negotiated enterprise; and a shared repertoire of negotiable resources accumulated over time’

196

Gender and Politeness

(Wenger, 1998: 76, cited in Holmes and Meyerhoff, 1999: 175). Thus, each community will develop a range of linguistic behaviours which functions in slightly different ways to other communities of practice. However, we need to modify this notion of community of practice slightly, since although there may be broad agreement as to the norms operating within that group, there will also be different ‘takes’ on those norms, and gender may play a significant role here in determining what each participant views as appropriate. This notion of a community of practice is particularly important for thinking about the way that individuals develop a sense of their own gendered identity; because it is clear that individuals belong to a wide range of different communities with different norms, and they will have different positions within these groups (both dominant and peripheral). Thus, rather than describing a single gendered identity which correlates with one’s biological sex, it is possible within this model to analyse a range of gendered identities which will be activated and used strategically within particular communities of practice and which is subject to change. This more productive model of gender makes it more difficult to make global and hence abstract statements about women’s or men’s language; however, it does allow for variations within the categories ‘men’ and ‘women’, and allows for the possibility of contestation and change, whilst also acknowledging the force of hypothesised stereotyping and assumptions about linguistic community norms. As Eckert and McConnell-Ginet state: ‘An emphasis on talk as constitutive of gender draws attention away from a more serious investigation of the relations among language, gender and other components of social identity: it ignores the ways difference (or beliefs therein) function in constructing dominance relations. Gender can be thought of as a sex-based way of experiencing other social attributes like class, ethnicity or age (and also less obviously social qualities like ambition, athleticism and musicality)’ (Eckert and McConnell-Ginet, 1998: 488, 489). Thus, we do not need to lose sight of the way that hypotheses about stereotyping functions within communities, rather, the stereotypes of gender, race, and class difference will be more or less salient dependent upon the community of practice, and each community of practice may develop different positions in relation to these stereotypes (see Bucholtz, 1999b). It may also be the case that certain activities within those communities of practice might be

Theorising gender

197

coded or recognised as stereotypically masculine or feminine, and thus certain types of linguistic activity may be considered by males and females as appropriate or inappropriate within an interaction and sanctioned by the group as a whole. Perhaps what needs to be added to the notion of the community of practice is a wider notion of the social and an awareness of the pressure that institutions can exert on communities and individuals. Without this notion of the larger cultural group, it is difficult to move beyond the community of practice and explain why certain values and forms of behaviour are globally valued more than others. This is not to reinstitute such global notions as society or the state entirely which the notion of habitus and community of practice seem to force into the background, but to be aware that institutions such as the legal system, the school system, the church, and so on, and the media where our ‘common experience’ is mediated, nevertheless also exert pressure on individuals and suggest possible persuasive positions for them to adopt. The relationship between particular communities of practice and the wider social group is, of course, a very complex one and is more often characterised by conflict than by simple affirmation. When this new more complex theorisation of gender is extended to the analysis of linguistic politeness, it results in a move away from stereotypical assumptions about femininity that have dominated discussions of women’s use of politeness, in most of the standard analyses of gender and language from Lakoff (1975) through to Holmes (1995).20 It is clear that we need to acknowledge the extent to which the notion of ‘women’ is classed and raced, particularly when we are considering linguistic politeness. As I shall argue in the next chapter, politeness is already gendered, classed, and raced, so that stereotypically it bears a signature of middle-class, white, femininity and this trace lingers on in the way that individuals construct their own sense of what is appropriate for them and for others which will manifest itself in what they consider to be polite and impolite behaviour; the way they think of their behaviour as polite or impolite; in the way that they judge others’ utterances; and what they consider to be appropriate reactions to politeness and impoliteness. This stereotyped connection between gender and politeness leads to certain expectations by members of communities of practice about what linguistic behaviour they expect of women and men.

198

Gender and Politeness

Conclusions This chapter has thus attempted to show the necessity of a more complex model of gender for the analysis of women’s and men’s use and understanding of politeness. It is essential to move beyond polarising women and men as distinct groups and concentrate more on the way that people ‘do gender’ in particular communities of practice, hypothesising their own appropriate gender performance and set of interpretative frameworks from the stereotypes and actual performances that they think are available to them and to others. Notes 1. An earlier version of parts of the introductory section of this chapter will appear as sections of a chapter entitled ‘Language and gender’, in Mary Eagleton’s forthcoming edited collection: Feminist Theory, Blackwell. 2. This argument seems perverse, especially since over-emotionalism is one of many qualities which are associated stereotypically with femininity and, hence, with women. Here, in a case of reverse logic, emotional response is considered a positive attribute, in that it is associated with male behaviour. 3. Indeed, perhaps one of the most important aspects of this work is that women felt that they could comment on an interruption by a male interlocutor, and rather than dismissing such behaviour as solely due to the particular chauvinism of that individual, they could relate it to wider societal structures which made available to men privileged powerful positions, which it did not provide for women. 4. Although Tannen claims that men can also do ‘rapport talk’ and women may do ‘report talk’, she argues that, in general, women are concerned with establishing rapport and men with providing information and finding a place for themselves in a hierarchy. 5. Many have challenged the notion that children are now reared in singlesex groups; whilst play within schools is often in single-sex groups, most other activities, such as group work, is not (Harness Goodwin, 2001). Even friendship groups and team sports are not necessarily organised on the basis of sex within schools. Within the workplace, single-sex groups are now no longer common, and whilst there are obviously single-sex friendship groups, groups seem now to be more mixed sex. 6. Although it used to be argued by feminists that ‘sex’ was the term used to refer to the biological differences between women and men, and ‘gender’ was the socially constructed set of attributes which women and men learn to adopt in order to construct their gendered identities, in recent years, the term ‘gender’ has become more fuzzy in its usage. It is now the term used most generally in discussions of differences in male and female

Theorising gender

7.

8.

9.

10.

11.

12.

199

linguistic production, because it seems to acknowledge that male and female are not such easily demarcated categories, on biological grounds, and that masculine and feminine are often elided with the categories of male and female. It now seems to be being used within language and gender research to refer to sex differences as well. However, whilst this is a productive way of viewing the negotiations of lesbians with stereotypes of language style, it is important to note that this, like many other models of speech and identity, is concerned with the speaker’s intention in producing particular forms of speech and Queen does not comment on the way that this speech style is interpreted by others, both straight and lesbian. However, whilst Butler’s work is very productive, it must also be noted that her reliance on notions such as performativity, developed within Speech Act theory, falls prey to the difficulties which I outlined in chapter 1 concerning the way that the individual is related to social processes and institutions. It may be argued that, since power and masculinity are correlated (however complex that relation is), interactional power can only be achieved by using masculinist strategies in speech; however, one’s position within a speech community may be advanced by using a range of different strategies, including the seemingly more co-operative/rapport ones, depending on the community of practice. Competitive talk is not always valued by communities of practice, which may code it as too direct, bullying, and overbearing. Although this option of a strong linguistic performance is available to secretaries, not all of them opt to present themselves in the same way. Some may choose more deferential linguistic behaviour because of their investment in femininity, or because of their view of their status within the organisation, and because of the way that they are treated by others, both at work and at home. We need to consider the analysis of power beyond the powerful– powerless binary divide, as it is clear that power relations are enacted within everyday interaction and result in a constant shifting along a scale of power rather than a simple slotting into positions of power or powerlessness. In addition, if we move away from the binary notion of power, we can analyse more positively the position of those who challenge their positions in relation to power; as Skeggs suggests: ‘to challenge powerlessness does not mean that one automatically shifts into positions of power. It means that one is refusing to be seen as powerless’ (Skeggs, 1997: 11). If we assume that asymmetrical power relations determine different styles of linguistic performance, and that gender difference is enmeshed with power difference in intricate ways, then we should be led to believe that women speaking to men will produce different styles of language from men speaking to men, or women speaking to women. This

200

13.

14.

15.

16.

17.

18.

19.

Gender and Politeness assumption would only hold if there were a clear correlation between gender and power, which there is not. Women’s class position is still determined largely by her relation to men: her father or her husband. However, professional women’s class position seems to be determined largely by income and education rather than those of their parents or husband (Sargent, 1981; Crompton and Mann, 1986). Trudgill’s (1972) work seems to exemplify this position where workingclass women are considered to be aspiring to middle-class norms through their assessment of their use of prestige speech forms. However, James (1996) has argued that the use of prestige forms, or the assumption that one uses them, is considerably more complex than Trudgill claims. Even so, her analysis still seems to make the assumption that working-class women tend to use prestige forms more than do men, and there is little analysis of those women who choose to use non-prestige forms, and the factors which go into their choice of those forms. It should, however, be noted that discussions of dialect and accent are not in themselves discussions of class. The discussion of class in linguistic circles has been made very difficult since the furore about Basil Bernstein’s work on restricted and elaborated codes, which seemed to suggest that working-class children used a poorer form of language and implicitly therefore had lesser cognitive skills. Trudgill’s work (1972) did at least have the merit of demonstrating that for certain groups of working-class speakers use of dialect or accent associated with the working classes was of greater value than affiliation with middle-class norms and Received Pronunciation (RP). As Whelehan (1999) has shown, there is a range of positions, for example, on whether singers such as the Spice Girls and Madonna are positive role models for women, or whether Girl Power constitutes a form of acquiescence with patriarchal norms. When she had Ali G on her show in 2002, he used some very strong language at a time (7.00–9.00 a.m.), when children might have been listening. She therefore apologised extensively after he had left the show and again the next morning. However, as I noted in ch. 3, Shaw has shown that breaking rules is a way of accruing power to one’s self within contexts such as the House of Commons. She argues that women MPs tend to adhere to the formal rules of turn-taking and they therefore do not manage to gain the illegal turns for themselves which their male colleagues do (Shaw, 2002). In certain recording sessions which some of my undergraduate students undertook at the University of Loughborough, in 1993, this was clearly the case. Some of the male students whose interactions were recorded saw intimate speech situations as stereotypically feminine and therefore spent a great deal of the time drawing attention to the fact that they were being recorded and addressing sexist comments to the person who

Theorising gender

201

was recording the interaction. Also, Cameron (1997) has shown that single-sex heterosexual male groups may use this seemingly feminine speech setting of informal gossiping to co-construct their heterosexual masculinity against a supposed homosexual male Other. 20. However, in a 2002 conference paper at the International Gender and Language conference, Holmes seems to be modifying her (1996) position, arguing that women and men in a business environment cannot be so easily distinguished in relation to their use of positive and negative politeness.

5

Gender and politeness

Introduction Given the model of gender described in the last chapter, and given the model of linguistic politeness as described in chapters 2 and 3, it is difficult, if not impossible, simply to approach the relation between gender and politeness as a question of an investigation of the production, by individual men or women of a number of linguistic features which are assumed to be unequivocally polite or impolite. What I should like to do instead is to consider the complexity of the relationship between gender and politeness, so that the common-sense nature of gender and politeness and their relation to each other is troubled. Here, I aim to analyse the way that certain practices which are considered to be polite or impolite are, within particular communities of practice, stereotypically gendered. As I discussed in chapter 4, these stereotypes do not actually exist as such, but are hypothesised by particular speakers and hearers within communities of practices, on the basis of their representation by others, and are then negotiated with. It is this connection between gendering of practices and assessments of politeness and impoliteness which is of interest. These stereotypes of behaviour which are considered to be appropriate within particular contexts feed back into individual participants’ assessments of what is appropriate in terms of their own behaviour. First, in this chapter, I analyse stereotypes of gender and politeness, and then move on to a discussion of the theoretical work on gender and politeness which I argue seems to replicate stereotypical views of women’s politeness, rather than describing women’s or men’s actual linguistic performance or interpretative frameworks. 202

Gender and politeness

203

I then consider two aspects of linguistic behaviour which have often been stereotypically associated with women: compliments and apologies; and I analyse two extracts from conversational data in order to challenge any simple view that women are necessarily always more polite than men.

Stereotypes of gender and politeness As I have argued throughout this book, at a stereotypical level, politeness is often considered to be a woman’s concern, in the sense that stereotypes of how women in general should behave are in fact rather a prototypical description of white, middle-class women’s behaviour in relation to politeness. The teaching and enforcement of ‘manners’ is often considered to be the preserve of women. Femininity, that set of varied and changing characteristics which have been rather arbitrarily associated with women in general, and which no woman could unequivocally adopt, has an association with politeness, self-effacement, weakness, vulnerability, and friendliness, as I showed in the last chapter. This manifests itself in the type of language practices which Lakoff described as ‘talking like a lady’ (Lakoff, 1975: 10). Rather than considering that this is an accurate representation of ‘women’s language’, this chapter will investigate other ways of thinking about the relationships between gender and politeness. What I should like to contest is the reifying of this view of the stereotypical behaviour of a group of women, and the extension of such a stereotype to all women. Women’s linguistic behaviour is often characterised as being concerned with co-operation (more positively polite than men) and avoidance of conflict (more negatively polite than men). This characterisation is based on the assumption that women are powerless and display their powerlessness in language; these forms of politeness are markers of their subordination. However, as I mentioned in the last chapter, stereotypes of gender have been contested for many years by feminists and have themselves been changed because of the changes in women’s participation in the public sphere. We can therefore no longer assume that everyone has the same ‘take’ on a stereotype, or that they share assumptions with others about what a particular stereotype consists of, or even that they accept stereotypes at face value rather than, for

204

Gender and Politeness

example, ridiculing them. Neither can we assume that certain forms of politeness are unequivocally powerful or powerless. Particularly in relation to gender stereotypes, politeness and impoliteness operate in different ways, so that impoliteness functions and signifies differently for certain groups of women. Behaviour such as swearing and directness, which might be considered impolite in certain linguistic communities, may often be excused or condoned for certain groups of men, particularly in the working classes, whereas for middle-class, particularly middle-aged, white women, it is may be judged to be aberrant. Indeed, the epitome of stereotypical language behaviour for males and females seems to be white, working-class men (direct, assertive, impolite) and white, middleclass women (polite, deferent, ‘nice’ to others). At a stereotypical level politeness is largely associated within Western countries with middle-class women’s behaviour. This may well be because politeness itself is generally considered a civilising force which mitigates the aggression of strangers and familiars towards one another and ideologically this civilising move is often associated with femininity.1 Masculinity, on the other hand, is stereotypically associated with directness and aggression. From the questionnaires and interviews which I carried out, middle-class, and working-class, white women themselves, particularly those who did not have paid employment outside the home, tended to find impoliteness of greater import than other groups, and slights and perceived lack of the appropriate level of politeness were often perceived by these women as a greater problem than for other groups.2 Although that is not to say that other groups of women or men do not find lack of politeness important (certain of the middle-class, white men I interviewed also stressed that they also valued politeness and found impoliteness difficult to deal with); however, white, middle-class women more than others focus on it as crucial in terms of assessing their everyday relations with others. This association between gender and class imperatives on politeness results in concerns about class separateness and gender coalescing. ‘Correct’ forms of behaviour for women, within assessments of politeness, particularly the more formalised views on politeness which appear in books on courtesy and etiquette, are also ‘correct’ forms of middle-class behaviour. Those elements which are judged to be not polite behaviour, that is examples of behaviour which are not acceptable in ‘polite (i.e. middle class) circles’, are

Gender and politeness

205

those which are considered ‘common’, ‘vulgar’, or which are not ‘decent’. These terms have a particular resonance for middle-class women and those working-class women aspiring to middle-class values. Thus, if we analyse some of the advice given in etiquette books, politeness is seen to be the avoidance of behaviour that is associated with the working classes: ‘never, never scrape the plate, unless you want to give the impression that you are normally ill-fed as well as ill-bred’ (Bolton and Bolton, n.d.: 22).3 Certain foods and drinks are associated with the working classes and thus, ‘fizzy soft drinks should not be offered as pre-dinner drinks’ and ‘it should not be necessary to say that bottled sauces should not appear on the table’ (Bolton and Bolton, n.d.: 94). This concern with avoiding stereotypically working-class behaviour is seen as avoiding behaviour which is ‘bad form’ and, as the reference to ‘ill-bred’ above clearly shows, polite behaviour is that which is normally associated with middleclass behaviour, which is itself characterised as ‘correct’ and ‘good form’ (Bolton and Bolton, n.d.: 73).4 Thus, politeness or concern for others is stereotypically associated with middle-class behaviour. Furthermore, politeness is often associated within English-speaking communities with being deferent, which Brown and Levinson have classified as negative politeness, often associated with powerlessness, and care for others, which is associated with stereotypes of femininity. Women’s linguistic behaviour, in many accounts, because it is seen as displaying powerlessness, is characterised as hesitant and unassertive and showing negative politeness for others through what is seen to be excessive use of respect and deference. These characteristics associated with deference and positions of unequal power become associated with the ‘natural’ behaviour of women and other subordinate groups, as Hochschild argues: ‘The deferential behaviour of servants and women – the encouraging smiles, the attentive listening, the appreciative laughter, the comments of affirmation, admiration or concern – comes to seem normal, even built into personality rather than inherent in the kinds of exchange that low-status people commonly enter into’ (Hochschild, cited in Scott, 1990: 28). Positive politeness is associated with being friendly and nice towards others, and because there is an association between white, middle-class women’s stereotypical behaviour and niceness, this form of politeness is then associated with women’s linguistic politeness in general.5

206

Gender and Politeness

Courtesy is a form of behaviour which is conventionally extended from white, middle-class men to white, middle-class women, and consists of behaviour such as opening doors for women, helping them on with their coats, walking on the outside on the pavement, giving up a place on the bus, and so on.6 This form of behaviour is anachronistic for many people in Britain, and the display of care for others within the public sphere is something which the Campaign for Courtesy, a pressure group which tries to encourage others to be more courteous, feels has deteriorated over the last twenty years (Gregory, 2001a).7 Even when courtesy was more accepted as a general standard of behaviour during the 1950s–1970s, it was characterised as a set of values which was contested or under threat because of the behaviour of certain groups or classes. Courtesy is generally subsumed in most people’s accounts under the general heading of politeness, especially for many older people; one book on etiquette from the 1970s states: Politeness between the sexes is particularly important. Both men and women err greatly in their behaviour towards one another. The attitude of many men is “they wanted equality, and they’ve got it”, and this is their excuse for letting a door slam in a woman’s face or failing to walk on the outside of the pavement when they are together, or paying any of the usual forms of courtesy. Women on the other hand, often adopt a rude and militant attitude towards men, and their behaviour when courtesy is shown them is sufficient to discourage even the most courteous male. Again, so many women neglect to return thanks for politeness, and behave so impertinently that any man might be forgiven for not bothering in the future. (Bolton and Bolton, n.d.: 34)

Courtesy is certainly something about which there is a considerable conflict between professional, middle-class women and professional men. At present it is very difficult to negotiate having courtesy extended to oneself as a woman without offending by refusing. For those men who open doors for women, courtesy may be intended as a display of care and respect, whereas for many feminists, it seems demeaning because it seems to suggest that the woman is incapable of opening doors, putting on her own jacket, and carrying her own bags, and it also seems to be non-reciprocal.8 For example, in a book on etiquette from the 1970s, the following sentiment is indicative of a wider set of advice to men involving displaying care for women: ‘Going up or down stairs, the rule is for the man to keep to

Gender and politeness

207

a lower level so that he may offer assistance if required; i.e. he follows the woman up a stairway, but precedes her downstairs’ (Bolton and Bolton, n.d.: 34). This characterisation of courtesy assumes that women need protecting by men. Thus, courtesy and etiquette, like other forms of politeness, can be seen to have particularly strong class and gender associations at a stereotypical level. Theoretical work on gender and politeness The major theoretical work on politeness and gender has been undertaken by Penelope Brown and Janet Homes, and I would like now to discuss their work in order to show how each of them draws on these stereotypes of women’s and men’s behaviour in relation to politeness. Penelope Brown in her work on the analysis of politeness among a Mayan community, argues that women in general are more polite than men (Brown, 1980, 1993). She states that ‘in most cultures women among women may have a tendency to use more elaborated positive politeness strategies than men do among men’ (Brown, 1980: 251). Her general model of politeness is one associated with care for others: ‘what politeness essentially consists in is a special way of treating people, saying and doing things in such a way as to take into account the other person’s feelings. On the whole that means that what one says politely will be less straightforward or more complicated than what one would say if one wasn’t taking the other person’s feelings into account’ (Brown, 1980: 114). This statement sees politeness as largely a matter of concern for others. She asserts that this greater use of positive politeness by women is due to power differences within this particular Mayan community, but that power differences can be seen to produce similar behaviour in other cultures: ‘men may assimilate more upper-class dignity and competition for power, while women, excluded from this arena, maintain solidary ties with one another’ (Brown, 1980: 114). However, as I shall show, her results are significantly more complex than this and seem to show just how difficult it is to prove that men as a group or women as a group use politeness in similar ways. She discusses the way that many linguists have concluded that women’s language tends to be more hypercorrect than men’s and hence more formal (Brown, 1980). As I mentioned in chapter 4,

208

Gender and Politeness

linguists such as Trudgill (1972) claim that this is because women tend to gain prestige through appearance and linguistic behaviour, since they cannot gain status through their job or income (Trudgill, 1972).9 The assumption that is made is that hypercorrectness and use of the prestige variety in English can be assumed to be markers of polite linguistic behaviour, and that this type of behaviour marks an unstable or insecure social position; thus Brown argues: ‘it seems reasonable to predict that women in general will speak more formally and more politely, since women are culturally relegated to a secondary status relative to men and since a higher level of politeness is expected from inferiors to superiors’ (Brown, 1980: 112). Here, Brown seems to be conflating politeness and negative politeness or deference, and she also seems to be assuming that an inferior social position will necessarily determine the type of language that is produced. However, she goes on to give the example of the Malagasy village studied by Elinor Keenan, where women’s speech is judged to be less polite than men’s, but in this instance, this type of speech is stigmatised by the society as a whole. Therefore, here stereotypes of women’s speech are assessed rather than women’s actual speech. Brown’s work focuses on speaker intentions and she does not concern herself overly with hearer interpretation and judgement, which is clearly crucial in concerns about status in this community. Brown sees politeness as being concerned with questions of social standing and this she sees as being of great importance for women. For her, since relationships in general are fairly stable, politeness levels are also fairly predictable. If there is a shift in the level or type of politeness used, then we are to assume that there has been a change in respect, an increase in social distance or a change of a face threatening nature. She argues that therefore most fluctuations in politeness levels are due to the mitigation of a Face Threatening Act (FTA). ‘Given then a range of politeness levels over a wide range of kinds of acts, we can infer degrees of social closeness and degrees of relative power in relationships. Thus, politeness strategies are a complicated and highly sensitive index in speech of kinds of social relationship’ (Brown, 1980: 117). She goes on to ask why and how women are more polite than men and she suggests that ‘women are either (1) generally speaking to superiors, (2) generally speaking to socially distant persons, or (3) involved in more face-threatening acts, or have a higher assessment than men have of what counts as

Gender and politeness

209

imposition’ (Brown, 1980: 117). Whilst this may be correct on a stereotypical level, I would argue that in fact in relation to women’s linguistic behaviour as a whole, these assertions do not necessarily hold. In Brown’s analysis of strengthening and weakening particles in Tenajapa, she asserts that the particles which weaken an utterance, hedging on its epistemic value, can be seen to play a role in negative politeness, and those particles which strengthen an utterance can be seen to play a role in positive politeness. As I have shown in earlier chapters of this book, making this assumption is fraught with difficulties, since the interpretation of the function of particular language items can only be understood within a particular context, judged within the framework of the hypothesised norms of a particular community of practice. However, Brown asserts that in this Mayan community, which is very clearly sex-differentiated, women use more strengthening particles when speaking to women (and to men) and they also use more weakening particles when speaking to men. Women speaking to women use more particles in general than men use to men. This is an important finding, since Brown shows that she is aware that women do not have a simple general style which all of them use in all circumstances; rather, their choice in terms of the use of these particles depends on the assessment of context and audience. Furthermore, when making this generalisation, Brown is aware of the importance of topic and relative knowledge in relation to the number of these particles which are included within speech, and the fact that when value judgements are given it is expected that these particles occur more than when evaluations are not being made. However, despite these provisos, she still asserts that in general women’s language use is more polite than men’s. She gives, as a case study, the use of the particle ‘ala’ (a little) which she asserts is a diminutive which seems to her to be functioning both as a negative politeness element when used by males in her examples, but also as a positive politeness element when used by females in her examples. Despite this awareness of the fact that ‘ala’ functions differently for different groups, she asserts that the use of ‘ala’ is seen to be characteristic of women’s speech. Holmes, whose work I discuss later in this chapter, comments on Brown’s analysis of these particles: ‘the particles she examined tend to occur most frequently in

210

Gender and Politeness

speech expressing feelings and attitudes, and . . . in her data women spent more time talking about feelings and attitudes towards events than men. It seems possible that the association of particular linguistic devices with women’s speech may reflect the fact that they occur more often in discourse types favoured by women’ (Holmes, 1995: 110). Thus, Holmes seems to be viewing women’s language at a stereotypical level, in that she assumes that women more often talk about emotions than men. As I discussed in chapter 3, the assumption that women’s speech is necessarily different from that of men often leads us to draw on stereotypes of feminine behaviour rather than on women’s actual linguistic behaviour. In general, even though there are several cases where Brown’s hypothesis is not proven when tested against her data, she still asserts that women and men’s speech differ significantly in relation to politeness use. Her results ‘contradict our initial impressionistic hypothesis, that women are positively polite to women, and negatively polite to men. Rather the data suggest that women are overall more sensitive to possibly face-threatening material in their speech, and hence use negative politeness to women as well as men, and are more sensitive to positive face wants and hence use positive politeness to men as well as women’ (Brown, 1980: 129). She suggests that this negative politeness use between women is due to the fact that ‘there is not a dichotomising of the social world into men vs. women . . . but that overall women are paying more attention to face redress than men are’ (Brown, 1980: 131). Thus, in her conclusions, Brown is aware of the complex interaction between social status and gender, and is also aware that politeness markers (if they can indeed be classified as such) have a range of different functions for those of different status and gender. However, because of stereotypical assumptions about the homogeneity of these speech communities, she is unable to challenge the notion that women’s speech will necessarily be speech that displays deference. Because of her concerns to present these Mayan women as homogeneous in terms of their speech, she notes that despite the clear differences that she has described in sex-roles (men beat women but women do not beat men; men use direct orders in public but women do not) in fact, this community is fairly egalitarian and women are not totally powerless, since they play a major role in the community both economically and socially. She herself notes

Gender and politeness

211

that the women’s speech community in Tenejapa is not uniformly powerless, for example, that women over childbearing age tend to speak more in line with the norms for men and some of them manage to achieve some powerful positions within the community. She argues that in communities where women have no social esteem, for example, low-caste women in India, their linguistic behaviour in relation to negative politeness will be different, ranging from highly honorific language to bald on record, since Brown argues that these people are treated as if they have no face. However, if politeness use is so closely linked to questions of status, those women who have some status in the community should display this in their language use, and there should be more of a sense of diversity within women’s language use within Brown’s account. Brown refines her position even more in her 1993 article on ‘Gender, politeness and confrontation on Tenejapa’, where she argues that within particular contexts, in this case the courtroom setting, women may use the speech forms most associated with feminine politeness sarcastically in order to perform FTAs. Whilst still affirming that women are more self-deprecating than men within the public sphere generally, and more positively polite than men, she analyses the way that women in a courtroom confrontation draw on these resources of feminine behaviour in order to behave aggressively to other women, ‘despite the strong constraints against public displays of anger, there is an institutionalised context and mode for confrontation: a dramatised outrage played against the backdrop of appropriate norms for female behaviour’ (Brown, 1993: 137). Thus even when women are not being polite, since they are being aggressive, they do this by drawing on indirectness and politeness ironically. Her conclusions from this work are that deference prevails when people are vulnerable within a society; thus, women in such a position will use more negative politeness. Positive politeness prevails if and when networks involve multiplex relations, where relations are multistranded. Cameron comments: Brown’s argument, however, is not that politeness works differently for men and women. It is that while both sexes must make the same calculations about the same variables (e.g. social distance, relative status, degree of face-threatening inherent in a communicative act), the different social positioning of men and women make them assign different values to those

212

Gender and Politeness

variables, and therefore behave differently. If Brown had explained the women’s ‘more polite’ behaviour as a simple consequence of either their feminine gender or their powerlessness she would not have been able to explain the fact that they are differently polite to male and female interlocutors (if it were only femininity, why should there be any difference? If it were only powerlessness, why be polite to your equals – other women – at all?). (Cameron, 1998a: 444)

Whilst I accept that in this way Brown’s work integrates a certain element of heterogeneity within her notion of women’s speech, she still characterises women as essentially powerless, whilst showing at the same time that within this community, there are women who exercise greater interactional power, within particular contexts. A similar finding can be seen in Smith-Hefner’s (1988) analysis of the use of polite forms in Java, where she notes that different cultures have different definitions of what counts as polite; she argues that in Java polite forms are associated with high status and with linguistic control and skill. ‘Where the register is linguistically complex and not everyone is able to use it effectively, mastery of the register may thus identity the speaker as distinguished or socially exemplary . . . Under such circumstances, we should expect that control of and access to these polite codes would show a pattern of differential distribution among speech community members often in relation to variation by class and gender’ (Smith-Hefner, 1988: 537). Because of this association of politeness with high status and verbal skill, although Javanese women are expected to be more polite than men within the family and this use signals their subservience (an assertion which seems to support Brown’s assertion about Tenejapan women), outside the family within the public domain politeness is associated with males. Brown and Levinson devote a section of their revised 1987 introduction to the question of gender and here Brown’s argument is extended still further so that gender becomes an even more complex variable. They argue: ‘empirical tests of Lakoff’s specific claims (that women are more polite than men) have by and large failed to substantiate them in detail . . . but the argument that women have a distinctive “style”, due to their distinctive position in society, is still being actively pursued, despite the persistence of negative evidence (no clear sex differences found) in much of the research’ (Brown and

Gender and politeness

213

Levinson, 1987: 30). They assert that rather than simply analysing data for sex differences: ‘in trying to understand the often very elusive and subtle differences between the language use of men and women we need to be crystal clear about exactly where and how the differences are supposed to manifest themselves’ (Brown and Levinson, 1987: 30). For example, we need to be clear about whether we are examining differences due to the gender of the speaker or the hearer or both. Of great importance here is the assertion that simply analysing data for gender difference is not adequate, since ‘we need constantly to remember the obvious but always pertinent fact that gender is just one of the relevant parameters in any situation, and is indeed potentially irrelevant in a particular situation’ (Brown and Levinson, 1987: 30). They are aware of the difficult relationship between gender and the other social variables which they examine in their work. For them, it is difficult to assess whether gender is at work; if we assume that gender and power (P) are the same, since all women are powerless, then we shall also have to take into account social distance (D) in relation to gender: ‘unicausal explanations in terms of P (that women are universally subordinate to men and therefore more polite) will not do justice to the complexities’ (Brown and Levinson, 1987: 30). Thus, although Brown and Levinson try to question the assumption that women are necessarily more polite than men, their data in general seem to prove that in most circumstances this is indeed the case. This is partly because their work focuses on speaker intentions and therefore can only deal with intentions to be polite rather than stereotypical assessments of politeness by others, which may be at odds with those intentions. Drawing on Brown and Levinson’s work, Janet Holmes argues that in general women are more polite than men. Her empirical studies belong to the ‘difference’ model of women’s language within feminist linguistics, influenced by Coates’ (1996) and Tannen’s (1991) work on co-operative and competitive strategies. Thus, Holmes asserts that women are more polite than men, as they are more concerned with the affective rather than the referential aspect of utterances since ‘politeness is an expression of concern for the feelings of others’ (Holmes, 1995: 4). Holmes states that she uses a broad definition of politeness, following Brown and Levinson, so

214

Gender and Politeness

that politeness refers to ‘behaviour which actively expresses positive concern for others, as well as non-imposing distancing behaviour’ (Holmes, 1995: 5). She suggests that women are more likely to use positive politeness than men; thus for her: ‘women’s utterances show evidence of concern for the feelings of the people they are talking to more often and more explicitly than men’s do’ (Holmes, 1995: 6). I aim to contest Holmes’ notion that women are globally more polite than men, arguing that this is in fact based on a stereotypical view of women’s language. For some women, this stereotype may be important, but for others it may be something which they actively resist and reject. What is important here is the sense of the variety in the hypothesisation of the stereotype and variety in the response to that stereotype in terms of what behaviour is then considered to be appropriate. Holmes attempts to tackle the question of whether women and men are polite in different ways. When she poses the question ‘are women more polite then men?’, she answers ‘it depends what you mean by politeness, and it depends which men and which women you are comparing, and it also depends on the context in which they are talking’ (Holmes, 1995: 1). However, this focus on the context-specific is frequently dispensed with in her work, in order to make larger generalisations. Holmes tries to suggest that there are global similarities among women; thus, she asserts that women generally are more likely to be verbally fluent earlier, they are less likely to suffer from reading disabilities and aphasia, but, perhaps more importantly, she asserts that women have a different attitude to language use from that of men: ‘Most women enjoy talk and regard talking as an important means of keeping in touch, especially with friends and intimates. They use language to establish, nurture and develop personal relationships. Men tend to see language more as a tool for obtaining and conveying information. They see talk as a means to an end’ (Holmes, 1995: 2). This is very similar to the position advocated by other ‘difference’ feminist linguists, who claim that women and men are brought up in differently gendered sub-cultures and thus use language in fundamentally different ways to achieve different ends (Coates, 1996; Tannen, 1991). Holmes bases her claims on extensive quantitative data, unlike many researchers in this field. However, even though she analyses a great deal of data, it is largely from white, middle-class

Gender and politeness

215

New Zealand women and men.10 Holmes is aware of the difficulties entailed in making generalisations about women’s speech from these data: many of the statements about the way women and men use language will inevitably appear as gross generalisations from specific studies of particular men and women in particular situations who belong to specific cultures, social classes, age groups, occupational groups, and so on. It would be possible to qualify every statement with this kind of detailed information, but it would also be tedious in the extreme. In discussing the research evidence every reference to women and men should be appropriately qualified, but for obvious reasons I have tended to refer to women and men. This is not intended to mislead or over-generalise. (Holmes, 1995: 29)

Throughout her work, there is a problematic tension between the specificity of her interpretation of her data and her wish to make generalisations from those data. The problem with generalisation is that it often reflects stereotypical views or our notion of the average, and does not reflect the behaviour of any real speaker. There is also a tension between the wish to make generalisations about the language behaviour of all men and all women, and recognising that there are differences in power among women: ‘in communities where women are powerless members of subordinate groups, they are likely to be more linguistically polite than the men who are in control’ (Holmes, 1995: 8). One can presume that in communities where women are not powerless, Holmes might argue that they therefore would not be as linguistically polite as powerless women. After this acknowledgement of the importance of context, she goes on to state, however, that ‘when all the necessary reservations and qualifications have been taken into account, I think the answer is “yes, women are more polite than men”’(Holmes, 1995: 8). Holmes associates positive politeness with women’s speech. In discussing positive politeness, she states, ‘women are much more likely than men to express positive politeness or friendliness in the way they use language’ (Holmes, 1995: 6). Positive politeness is here seen to be synonymous with friendliness, and seems part of a general stereotype about the way that women should behave. However, even she recognises that distinguishing between positive and negative politeness is difficult: ‘in fact, there are few speech acts which are intrinsically negative politeness speech acts. Linguistically expressed negative politeness generally takes the form of expressions

216

Gender and Politeness

or strategies which reduce the effect of face threatening speech acts’ (Holmes, 1995: 154). In discussing the positive politeness strategies which New Zealand women use as reported by Pilkington, Holmes states: for the women, being negatively polite involves avoiding disagreement. Being positively polite is being friendly, and this involves confirming, agreeing and encouraging the contribution of others. But these politeness strategies are not typical of the interchanges described . . . between males. These young New Zealand men . . . are quite prepared to disagree baldly and to challenge the statements of others overtly. Indeed, for this group, insults and abuse appear to be strategies for expressing solidarity and mateship, or ways of maintaining and reinforcing social relationships. (Holmes, 1995: 66)

However, here there seems to be little difference in effect in what each of the groups is doing: the New Zealand women are expressing solidarity in much the same way as men but they are using different strategies. As Bergvall, Bing, and Freed have argued, once a researcher decides to analyse sex difference in language, they are forced to concentrate on difference alone and this evidence is used to argue that male or female behaviour are fundamentally different (Bergvall et al., 1996). This conflict between the wish to generalise and the awareness of the specificity of the material arises throughout her analysis. For example, there is a conflict between allowing for the multivalence of linguistic elements and the need to do quantitative analysis, as can be seen in the discussion of tag-questions, where she notes, in a rather tautological fashion, that ‘men generally use canonical tag questions more often than woman do to express uncertainty and ask for confirmation, while women use tag questions more often than men in their facilitative positive politeness function’ (Holmes, 1995: 85). These categories are in themselves judgemental rather than analytical. This problem also arises when Holmes then goes on to examine the use of facilitative tags by those in positions of power, since she argues that ‘powerful participants used more of the facilitative tags’ (Holmes, 1995: 85). If she is arguing that both the powerful and women use facilitative tags more, then we should be led to assume that there is a correlation between these two groups. The problem seems to have arisen because of the use of this judgemental categorisation of tags, since ‘facilitative’ is already an evaluation of the speech being analysed. This problem occurs throughout Holmes’

Gender and politeness

217

work in that evaluative categories are used, and the evaluations of the participants are not considered. Holmes attempts to move analysis to an examination of functions rather than forms; however, when she finds that men and women use a form to the same extent, she interprets their use differently, according to its function: in the use of ‘I think’, for example, she found no difference between its use by men and women in her samples, ‘but there was a contrast in the predominant functions for which it was used by women and men. Women used I think as a booster more frequently than as a hedge . . . whilst the reverse was true for men . . . and women used I think more frequently than men as a politeness device, especially as a positive politeness device, boosting an utterance expressing agreement with the addressee’ (Holmes, 1995: 94). This interpretation of the data attributes an unproblematic function to the use of ‘I think’ which may not be shared by the participants. Again, with the analysis of ‘sort of’ she states: although there was little or no difference in the data analysed in the total number of instances of these forms used by New Zealand women and men, in general women used them as politeness devices more often than men, and men used them more often than women as epistemic devices in their referential function. This pattern of female concern with affective meaning and male with referential meaning . . . illustrates once again the different orientations of women and men in interaction. (Holmes, 1995: 96)

However, this seems to be an evaluation of women’s speech, and these stereotypical views of women’s speech as more polite seem to be influencing the interpretation of the data. Furthermore, if data do not prove Holmes’ initial hypothesis, then she interprets the data differently in order to make them fit; for example, in an experiment she found that boys and girls use interruption in supportive ways equally. She states that ‘it is interesting to speculate on why the usual pattern of disruptive interruptions by males was not evident in this data’ (Holmes, 1995: 54). She then indeed speculates on why boys and girls behave in similar ways in this context, for example, that they were working in pairs and hence co-operative behaviour was demanded by the task and context; or that the males had not as yet learned male-dominance patterns. In another instance, in examining comments made after presentations by females and males, she divides the comments into supportive, critical, and antagonistic. Males and females were equally

218

Gender and Politeness

supportive; males were more antagonistic than women (which she discusses in greater detail than the other results); but the higher level of women being critical of the presentations she dismisses as determined by the context, ‘where it was clear that criticism had been invited’ (Holmes, 1995: 47). She then goes on to make a generalisation from this that ‘overall then this detailed analysis . . . provides further support for the suggestion that New Zealand women tend to be more sensitive to the positive face needs of their addressees than New Zealand men (Holmes, 1995: 47). In both of these cases, when data do not fit in with her overall hypothesis of difference, it can be dismissed because of context-dependent imperatives. Holmes interprets utterances as polite largely on a grammatical basis, even though she recognises that each lexical item has multiple possible interpretations and that there is no clear unproblematic link between form and function. Holmes focuses on grammatical features such as questions, as she feels that these indicate politeness; she suggests that who asks questions most is important in gender terms, and she goes on to state that questions are one way of handing the floor over to another speaker (Holmes, 1995: 31). She is aware that questions and interruptions may be differently judged by hearers: ‘what is perceived as rude, disruptive and impolite by women, may be acceptable and normal in male interaction. And when women politely (according to their norms) avoid interrupting others, they may be interpreted by males as being reluctant to get involved, or as having nothing to say’ (Holmes, 1995: 53). Thus, what her data are clearly showing is that politeness is an evaluation of behaviour rather than an intrinsic grammatical quality, but her model of analysis, based as it is on Brown and Levinson’s, forces her into a grammatical- and speech-act based analysis. Holmes interprets her data according to stereotypes of female and male. Despite the fact that her book was published relatively recently, Holmes makes certain rather outdated assumptions about women and the private sphere, for example: the amount of talk contributed by women and men differs in public, formal contexts, compared to private, intimate contexts . . . men tend to value public, referentially orientated talk, whilst women value and enjoy intimate, affectively orientated talk. Each gender may be contributing more in the situation in which they feel most comfortable. Correspondingly, women may experience formal public contexts as more face threatening

Gender and politeness

219

than men do, while men perhaps find private and intimate contexts less comfortable. Each gender contributes least in the situation they find less comfortable. (Holmes, 1995: 37)

This may, in fact, be an accurate assessment of some women’s and men’s responses to stereotypes of what is appropriate for each sex, but it is important to see this as a stereotype rather than to assume that it is a reflection of actual behaviour.11 In order to examine these problems in more concrete terms, I now analyse two of Holmes’ concerns in her book, particularly as these are ones which are most associated with the notion of women’s speech, and which seem to be most stereotypically associated with positive and negative politeness: compliments and apologies.

Compliments Holmes argues that ‘a compliment is a speech act which explicitly or implicitly attributes credit to someone other than the speaker, usually the person addressed, for some “good” (possession, characteristic, skill, etc.) which is positively valued by the speaker and the hearer’, and she also asserts that ‘compliments are generally regarded as the paradigm of a positive politeness strategy’ (Holmes, 1995: 144; Holmes, 1986, cited in Holmes, 1995: 117). Therefore, a compliment for Holmes functions as an unequivocal marker of positive politeness. But compliments can also function in very different ways depending on the context: compliments can set up a debt, or they may be intended to be interpreted negatively. Holmes is unable to see compliments as anything other than positive, but she is willing to admit that ‘very clearly, the relationship between the complimenter and recipient is crucial in accurately interpreting the potential functions of a compliment’ (Holmes, 1995: 118). She does not analyse these relationships, however. She is aware that context plays a role in whether a compliment is interpreted as an aspect of positive politeness or not, but she is forced to assume that compliments can be interpreted globally, because of her concern with generalisation and, influenced by Brown and Levinson, universal statements. Thus, whilst she admits that: ‘in some relationships compliments will be unwelcome because they are experienced as ways in which the speaker is asserting superiority’, her general positive model of

220

Gender and Politeness

compliments holds (Holmes, 1995: 119). She lists the functions of compliments – ironic, sarcastic, flattery, patronising, expressive of solidarity, praise, envy, or admiration – but, despite this very mixed list, in her data she analyses only those compliments which function to express admiration. She asserts that ‘provided it is not sarcastic, a compliment on someone’s appearance such as “you’re looking wonderful” is difficult to interpret as anything other than a positively polite utterance’ (Holmes, 1995: 131). However, there may be a range of different motivations and interpretations for such a remark: for example, the hearer might consider that the speaker is being insincere and is only complimenting because he/she wants something – i.e. that it is serving some longer term goal; or it might be interpreted as suggesting that the person does not look good at all, but that the speaker is being kind; or it might be interpreted as insinuating that the person looks better than they did before, because they looked dreadful before. In her data she does not include street remarks, where there is a gendered divide between intention and interpretation, and there is also a difference in the hypothesised intention – some women see them as compliments, some as harassment; some men assume that their remarks are intended to be positive, others may see them as a way of expressing hostility towards women without using more openly aggressive language, or as a way of showing their solidarity with other men in their community of practice. Despite this diversity of function of compliments, and clear disagreements about the interpretation of compliments, Holmes produces a total number of the compliments in her data, sorted by gender, and asserts that ‘complimenting appears to be a speech event occurring much more frequently in interactions involving women than men’ (Holmes, 1995: 122). On some occasions, she seems to recognise that compliments have different functions and interpretations, as ‘women may regard compliments as primarily positively affective speech acts, for instance, expressing solidarity and positive politeness, while men may give greater weight to their referential meaning, as evaluative judgements, or to . . . potentially negative face-threatening features’ (Holmes, 1995: 123). This is a key interpretation since she has shown how many various interpretations there are, even though she has globally assigned certain types of compliment to each sex. Although her model of politeness is focused on speaker intention, hearer interpretation necessarily asserts itself, for example, she states, ‘it is possible that men more than

Gender and politeness

221

women more readily perceive compliments as face-threatening acts’ (Holmes, 1995: 124). Holmes then tries to rationalise her assertion that globally women use compliments more than men, for, she argues that ‘compliments between women are most frequent . . . but men compliment women more often than they compliment other men’ (Holmes, 1995: 125). In this way, she, like Brown (1980, 1993), is able to see that the two sexes do not always operate in globally different ways but that there may be patterns to the way that the sexes behave according to context and the sex of the interactants. Rather than simply assuming that this difference is part of women’s nature, she asserts that perhaps women are complimented more than men because ‘they know women value them’ (Holmes, 1995: 125). She also suggests that women are complimented more, because, like children, they are subordinate. For Holmes, men and women compliment in very different ways globally; women tend to compliment on appearance and men focus on possessions.12 She draws on stereotypes of males being more able to risk the interpretation of their speech as face threatening when she asserts that ‘compliments on possessions . . . are much more vulnerable to interpretations as face-threatening acts . . . since there is the possibility that the complimenter will be heard as expressing desire for or envy of the object referred to’ (Holmes, 1995: 131). However, there is a similar risk of face threat when in the supposedly feminine strategy of complimenting on appearance, where a compliment such as ‘you’re looking very smart today’ might be taken as implying that you do not normally look smart. Compliments are extremely risky for both males and females as it is very easy for them to be misinterpreted; whilst they may be intended as positive politeness, they may be interpreted as face-threatening, for example, as overly familiar, intrusive, and impolite. Because Holmes does not consider the responses to compliments, whether they are accepted, deflected, or challenged, in any detail, she is unable to claim to be able to know that the linguistic behaviour she categorises as compliments functions as such for the hearers. Apologies At the level of stereotype, apologies are often associated with feminine behaviour, with excessive apologising, particularly the use of

222

Gender and Politeness

‘sorry’ and self-deprecation being assumed to be characteristic of women’s behaviour. However, apologies cannot be considered to be a formal linguistic entity, as I noted in chapter 2, since they can be made using a wide range of different linguistic strategies. Because of this diversity, it is possible to misinterpret apologies and assume that someone is apologising when they are not; to assume that someone apologising is, in fact, insincere; or to overlook an apology which is phrased in an indirect way. Thus, quantitative analysis will analyse only those elements which the analyst can recognise as clearly constituting an apology, such as those containing the words ‘sorry’ or ‘I apologise’. Holmes claims that women give and are given apologies in different ways from men. In her comparison of the numbers of apologies given and received in her data, she claims that women gave 75 percent of the apologies and received 73 percent. However, with this global picture of women as apologising more and receiving more apologies, we have to exercise some caution. If we assume that those in positions of subordination apologise more, and that women are in such a position in relation to men, then we should expect more apologies from women to men because of these power differentials. She asserts: ‘we are likely to apologise to those who are more powerful. And we are likely to apologise more profusely and extensively if the offence is serious’ (Holmes, 1995: 174). However, in fact, her data suggest otherwise: ‘men apologise twice as often to women as they did to men, regardless of the women’s position in relation to the apologiser. Interpretation must be speculative, but for men perhaps it is easier to apologise even to a woman boss or social superior than to a man’ (Holmes, 1995: 175). If this is the case, that men apologise more to women than to men, then we have to revise the hypothesis that subordinates always apologise more to their superiors, since the model of power which Holmes employs cannot deal with the position of the female boss, who, strictly speaking within Holmes’ terms, would have to be seen by the male employer as a subordinate because of her gender. The statistics given by Holmes do not back up her original hypothesis that women apologise more than men, and the raw data suggest there is no real difference between men and women except that which you would expect through randomness: she notes: ‘the resolution of this puzzle will involve exploring other social features of apology . . . part of the answer may lie in differential

Gender and politeness

223

perceptions by women and men of verbal politeness devices . . . women may regard explicit apologies for offences as more important in maintaining relationships . . . than men do’ (Holmes, 1995: 159). Thus Holmes, in order to try to make her data ‘fit’ her hypothesis, argues that women will tend to use clear apologies whilst men do not. She argues that men tend not to use more formal linguistic strategies ‘since formal linguistic strategies are generally not considered appropriate between people who are close friends’ (Holmes, 1995: 162).13 Again, as with her analysis of compliments, whilst employing a model of politeness which focuses on the speaker’s intentions, she is forced to try to interpret her data by calling on notions of the experience or evaluation of hearers and speakers: she argues, ‘it seems likely that while apologies may be experienced as admissions of inadequacy by men, that is they emphasise power differences, they are regarded by women primarily as ways of restoring social harmony and expressing concern for the other person. Additionally, it may be the case that the society as whole, both women and men, recognise the high priority that women place on politeness strategies as interactive tokens’ (Holmes, 1995: 176). Furthermore, she asserts that ‘failing to apologise to a woman is likely to cause greater offence than overlooking the need to apologise to a man’ (Holmes, 1995: 208). Thus, Holmes’ analysis, whilst based on the analysis of speaker intentions, finds that analysis of the differential judgement of the impact of apologies on hearers is necessary. However, she does not have the means to call on the judgements of the interactants apart from her own intuitions and stereotypes which she assumes the hearers are drawing on. Holmes is aware that stereotypes of gender-appropriate behaviour inform the use of apologies, as she states: ‘what society calls polite linguistic behaviour is largely based on women’s norms of interaction . . . consideration for the feelings of others has been the hallmark of every aspect of women’s verbal behaviour which has been examined’ (Holmes, 1995: 194). However, rather than recognising that this is in fact a stereotype, she assumes that ‘women’s norms’ exist in some tangible form. She seems to be trying to reclaim these ‘women’s norms’ from the stigma and ridicule as ‘redundant verbal frippery’ which often surrounds them, arguing in fact that women’s speech is not powerless but surprisingly influential: ‘despite their lack of social power, women have considerable

224

Gender and Politeness

social influence: their linguistic behaviour determines the overt and publicly recognised norms of polite verbal interactions in the community’ (Holmes, 1995: 194). Evaluation of ‘men’s’ linguistic behaviour begins to enter into her account towards the end of the book, so that ‘polite verbal interaction is based on women’s talk, and it has a very limited place in the male public sphere. It is not that men cannot do it. Rather it seems that most of the time they choose not to’ (Holmes, 1995: 195). She suggests that men denigrate polite behaviour since ‘polite behaviour is acceptable in contexts where nothing important is happening (in men’s perceptions). It has no place when important decisions are made’ (Holmes, 1995: 196). This suggests that those places where ‘nothing important is happening’ are associated with women. Because of her alignment with ‘difference’ models of women’s language, Holmes’ main aim is to change a negative evaluation – women’s language is weak – to a more positive evaluation, and to show that the same language items used by men and women can be judged differently: one (female) person’s hedge may be another (male) person’s perspicacious qualification . . . the association of linguistic markers of tentativeness and a high incidence of epistemic modal devices with insecurity, lack of confidence, powerlessness, and subordinate status, is to a large extent restricted to studies of women’s language. Epistemic devices are not interpreted this way when used in scientific discourse (which is dominated by men). There they are regarded as evidence of judicious restraint and meticulous accuracy. (Holmes, 1995: 111)

Thus, when she asserts that women’s language is more polite, she is also implicitly and sometimes explicitly stating that women’s language is better than men’s language. Holmes seems to argue that ‘women’s norms’, that is, polite behaviour, lead to better interaction: ‘linguistic behaviour which follows women’s norms can result in better working relationships, better understandings of complex issues and better decision-making’ (Holmes, 1995: 198). She suggests that ‘high quality exploratory interaction is essentially collaborative’, rather than the conflict model which currently holds (Holmes, 1995: 212). She also asserts that ‘the quality of a discussion is likely to improve when women get a more equal share of the talking time’ (Holmes, 1995: 212), and ‘on average females are much better than males at providing a favourable context for the kind of talk which is likely to lead to better understanding and

Gender and politeness

225

cognitive progress’ (Holmes, 1995: 212, 217). She suggests that if women’s norms of politeness were followed in discussions, then there would be an improvement in interpersonal relations, understanding, and performance in general. However, she also argues that women’s self-effacing behaviour, that is, over-politeness, can often lead to some groups of women being disadvantaged (which, paradoxically, seems to accept the assumptions of the ‘male’ characterisation of politeness as being where ‘nothing important is happening’). She suggests that ‘female students are generally not getting their fair share of the talking time. They are too polite’ (Holmes, 1995: 199). She also suggests, rather paradoxically, that ‘language learners need to be informed about gender-appropriate ways of using pragmatic particles, and the ways women and men use speech acts such as compliments and apologies’ (Holmes, 1995: 208). However, this assertion needs to be questioned, since, if we are to assume that what she is arguing is that men can apologise just as much as women do and in fact should be encouraged to do so, then language-learners should be taught women’s norms rather than men’s. Because of this positive view of politeness and women’s speech in general, Holmes argues that men need to change: ‘many males need explicit practice in enhancing their conversational competence’ since ‘what has been called “polite” language . . . has also proved to be cognitively beneficial language. Linguistic politeness contributes to better understanding, and may assist people to reach better decisions. Finally interactions which involve the use of positive politeness strategies are generally pleasant and enjoyable experiences. Being polite makes others feel good. There are many sound reasons, then for recognising the value of polite speech’ (Holmes, 1995: 229). However, as I showed in chapter 2, politeness is not always interpreted positively and may be viewed as manipulative or excessively deferent and, hence, weak. Thus, whilst showing clearly that politeness is associated with women at a stereotypical level, I would argue that Holmes’ analysis does not show that women in general are more formally polite than are men, as she asserts, but merely illustrates the difficulties of a methodology which focuses on the intentions of speakers and assumes that politeness can be recognised objectively by the analysis of formal features. Her analysis also demonstrates the difficulties of a model of gender which assumes that men and women are necessarily

226

Gender and Politeness

different and that they conform in their linguistic behaviour to gender stereotypes. Analysis of gender and politeness In analysing politeness in relation to gender, I would argue that we need to consider how to analyse the way that hypothesised stereotypes of gender are drawn upon by speakers and hearers in order for them to try and work out what are appropriate forms of behaviour. We as analysts also need to be aware that only the judgements of the speakers and hearers about what constitutes polite or impolite behaviour can lead us to the description of polite and impolite utterances. We need, therefore, to be aware that utterances may have a range of different interpretations. Kharraki’s (2001) analysis of Moroccan women and men’s bargaining in Arabic dialect illustrates that it is possible to analyse gender difference without assuming that women are more polite than men. He shows that women in Eastern Morocco, when bargaining with shopkeepers, drive just as hard a bargain as do men, but occasionally they draw on different strategies. Thus, they may be judged by the shopkeeper to be just as polite or impolite as men, but perhaps restricting themselves to forms of behaviour which are stereotypically classified as more feminine. For example, in one extract from his data, a woman says to a man who is selling onions: W: G: W: G:

How much are these onions Hassan? 60 doro Oh ! Don’t send your customers away! Reduce the price a little! On Monday they only cost 40 doro There is not much profit in it hajja. [respectful term for older Moroccan female] (Kharraki, 2001: 623)

In another example, he shows that women use very many strategies when bargaining, calling on the shopkeeper’s pity: W: G: W:

G:

Those potatoes look good, my fine fellow. How much do you want for them first? 80 doro. They’re very good quality. We’re not criticising the quality. Just see how you treat us. By God, if there were no potatoes, we and our children would die. We are very poor. God help you. Just choose some potatoes. Take it easy. (Kharraki, 2001: 625)

Gender and politeness

227

Because there are certain parameters within which bargaining is successfully executed, these seemingly quite forceful tactics on the part of women, insulting the shopkeeper and suggesting that he is trying to starve the women and her children, are not considered impolite by the shopkeeper. The interactional power of these women contrasts quite markedly with the stereotypes which many Westerners have of passive, deferent women within the Arab world. To illustrate the sort of analysis which focuses on gender and politeness without assuming that everyone behaves according to stereotypes of feminine behaviour, we might consider the following extract. Here, a New Zealand white, middle-class, middle-aged woman [K] who is staying on holiday in the home of a white, middle-class, younger, British woman [D] at the same time as D’s mother [M], a white, working-class, middle-aged woman, is visiting. K has been travelling around the world with her husband and is staying with D and her husband [F] because of a mutual acquaintance. D’s husband, a white, middle-class, young man, [F] is preparing lunch. The extract takes place in the dining room with all participants standing near the door. In this extract, which takes place just before the serving of lunch the day before K is about to leave, we can see that the context of the exchange rather than the force of feminine stereotyping determines to a great extent the degree to which the females here feel that they need to continue to thank K for a gift which she has given. We can also note that stereotypes may in fact inform some women’s habitual form of self-presentation, whilst not constituting the whole of that person’s identity: 1K: 2K: 3K: 4D: 5D: 6M: 7D: 8M: 9K: 10K: 11M: 12K: 13D:

D (.) here’s a little colourful Maori shell oh and I’m sorry we’re down to the ones that haven’t got nice bright colours in them(.) when you come to New Zealand you can come and pick your own off the rocks [ (laughs) just look at that= =beautiful= =isn’t that gorgeous [ that’s a real shell= =have you seen these ? that’s for you to take home if you’ve got room in your bag (gives shell to M) [ oh that’s lovely [ if not leave it behind how nice of you

228 →

Gender and Politeness 14M: 15D: 16K: 17D: 18K: 19M: 20D: 21M: 22K: 23K 24D: 25M: 26D:

[ oh that’s lovely=

=oh thanks that’s really lovely have you seen the pauwa shell before ?= =no do you know them ? lovely isn’t it ? never what(.)untreated are they K? no (.) well these have been polished (.) normally they’ve got a roughish back which looks like that there= =marvellous= → =oh that’s lovely thank you very much I love the colours= =aren’t they beautiful? [ → 27K: yes well this is I’m sorry but I’m sort of after a couple of years our bundle is well picked over and we’re down to the(.) that’s 28D: [(laughs)] 29K: got a nice inside though(.)the thing that makes them different the [ ] [ ] 30D: yes 31M: yes 32K: American abalone the American abalone has the opaly colours [ ] 33D: yeh 34K: in fact the one your mother has got is bluer than yours= 35D: =mmm I’ll just go and show F 36F: oh that’s beautiful (goes back to kitchen) 37M: and then the colours are in them and then they [ 38K: yes are then then if they’d been 39K: washed backwards and forwards in the tide they’re just like that [ ] 40M: yes yes 41K: but they’re usually broken= 42M: =oh I see= 43K: =so what you’ve got to do is to get the fresh really fresh shells and then 44K: grind the crusts off them it’s only a calcium(.) yes and they just lay on [ ] 45M: that’s really lovely 46K: our rocks= 47M: =oh well= 48K: =we just use them for ashtrays and things(.)but they’re unique [ ] 49M: mmm 50K: to New Zealand [ 51M: oh they are? → 52D: no I’ve never seen anything= 53M: =I’ve never seen anything like that (.) really lovely thank you very 54M: much [ → 55D: RIGHT well, I think the easiest thing to do is. if I start serving 56D: everyone. (Data: 56.5) EXAMPLE 11

Gender and politeness

229

If we analyse this exchange simply on the level of the number of politeness markers which are exhibited here, as Holmes and Brown have done, we might argue that this exchange between a group of women is characterised by a high number of positive politeness strategies on the part of D and M, mainly in the form of repeated thanks and by a fairly high number of negative politeness strategies, in the form of apologies, on the part of K. For example: →

25M: 26D:



27K: 28D: 29K:

=oh that’s lovely thank you very much I love the colours= =aren’t they beautiful? [ yes well this is I’m sorry but I’m sort of after a couple of years our bundle is well picked over and we’re down to the(.) that’s [(laughs)] got a nice inside though(.)the thing that makes them different the

In lines 25 and 26 M and D praise the shells in much the same way as they do later on in the interaction, for example: →

52D: 53M: 54M:

no I’ve never seen anything= =I’ve never seen anything like that (.) really lovely thank you very much

In this later example they use almost the same language as well as expressing the same sentiments. In the first example, K apologises for the poor quality of the shells. Since social distance is fairly low and the power relations are fairly evenly balanced, they might seem to be displaying a very feminine form of speech behaviour in thanking and apologising excessively. We might assume that the gift of two shells for putting someone up might merit perhaps one or two ‘thank you’s’, but here, D and M consistently and repeatedly praise the gifts and thank K (see lines, 5, 6, 7, 8, 11, 13, 14, 19, 24, 25, 26, 45, 52, 53). Since there seems to be a disparity between the gift and the degree to which D and M are thanking K, we can only assume, according to Holmes’ model that these women are behaving in a feminine over-polite way or that they are signalling a subordinate relationship to K. However, when I discussed this interaction with D and M, they stated that what they were trying to do was to get K to sit down so that they could start lunch. They felt that that by saying ‘thank you’ and praising the shell, they were closing down the conversation, allowing the topic to be changed to the subject of lunch, which, in line 55, D finally does. Instead of accepting the

230

Gender and Politeness

thanks and praise, K interprets their moves as a sign that they are interested in hearing more about the Pauwa shell in general and she then continues discussing the shells. D and M also stated that they found that K always talked a great deal and during her stay they had tried to develop strategies to cut short her talk. Thus, although here the strategies adopted might seem to be stereotypically feminine, because they seem to be over-polite, in fact the aim (although perhaps relatively unsuccessful since K continues to talk) was to terminate the discussion rather than to give an excessive amount of thanks. Furthermore, K might be seen, according to Holmes’ model, to be self-effacing, using negative politeness to minimise the value of the gift that she is giving to D and M; for example, K states in lines 1–3: 1K: 2K: 3K:

D (.) here’s a little colourful Maori shell oh and I’m sorry we’re down to the ones that haven’t got nice bright colours in them(.) when you come to New Zealand you can come and pick your own off the rocks

Brown suggests that the Mayan women in her data use diminutives more than men as a way of devaluing or limiting the claims that they make; here ‘little’ seems to devalue the shell, and she explicitly apologises for the lack of bright colours in the shells. Later in the exchange, as noted above, she again apologies when D praises the shell, saying ‘Yes, well this is I’m sorry but I’m sort of after a couple of years our bundle is well picked over’, referring to the fact that she has been travelling for several years around the world and the best shells that she brought with her as gifts have been given to others. She concedes that the one she has given to D ‘has a nice inside though’, again drawing attention to the fact that it is not a good shell in comparison to others. We might assume that apologising for gifts, when power relations are fairly equal, is an indication of the low self-esteem or excessive femininity of K. However, when compared to the assertiveness, as judged by D and M, of the rest of her speech, these apologies cannot be read as simply indicating that K’s speech is self-deprecating, particularly if we consider the difficulty remarked upon by D and M in getting her to stop speaking. Her speech is regarded by D and M in terms of her verbal habits, of speaking too much, rather than in terms of apologising too much, here. In fact, here we may be forced to analyse this instance of her speech style simply as an example of her trying to apologise for the fact that her

Gender and politeness

231

small gift does not equal the hospitality that she has received from D, F, and M (a point which both D and M themselves brought up when I interviewed them); and her apologies may be a response to what she sees as excessive thanking by M and D. Thus, rather than simply analysing data in terms of the way that males and females act differently in relation to their use of politeness, it is important to analyse what politeness, or what seems to be functioning as politeness, is being used to achieve within a particular community of practice. Here rather than seeing politeness as the sole production of one speaker, we need to see that the type of politeness used and the functions to which it is put are constructed and negotiated with in a concerted effort by all of the participants. In the next extract, if we were to use Holmes’ and Brown’s model of gender and politeness, we might again assume that the sex of the participants was necessarily significant in the interaction. In a misunderstanding which is then resolved, a group of young, white, working-class friends, a female employed as a bus driver [C], a male, her partner [T], who is unemployed, and another female [A], who is employed in another town and visiting her parents, discuss their plans for an evening out. The apology in the extract is something which is worked at jointly, even though the fault seems to lie largely with only one of the participants [A]. Thus, quantitative analyses of apologies which only analyse the production of explicit apologies by one speaker need to be questioned. Here the interactants are discussing meeting up with a group of friends at a pub: 1T: 2A: 3C: 4C: 5C:



6A: 7C: 8C: 9A: 10C: 11A 12T: 13A:

they might go for a curry after(.)that’s my bet anyway [ ] [ yeh but I haven’t found out yet what time I’m working tomorrow(.) so today I finished by about half past one quarter to two so the day sheet hadn’t been filled out(.) [ ] yeh I’m going to phone them up in a minute or when we go out we can drive down and I’ll jump out [ yeh is your car working?= =it’s it’s away at the moment = =um= =at the hospital [ so how can we drive round there?=

232

Gender and Politeness

=I thought you said you had your car I haven’t got a car here [ 16C: what did you say 17C: when she came in ? 18T: I said I presume you came by car 19A: [(laughs)] 20C: I thought you 21C: must have had your father’s car= 22A: =you you must be JOKING 23C: no when T said I presume you’ve got your 24C: car and you went yeh [ 25A: well why why was I saying the other day that I 26A: was going to hire a bicycle 27C: I would just have thought you had you 28C: might have had your father’s er [ 29A: no I er rather thought I might have my er 30A: father’s car but er no(.) it hasn’t even been suggested= 31C: =oh= 32A: =and I can’t ask= → 33C: =oh well this calls for a bit of phoning because it’s gonna have to be a 34C: bus up there cause it takes about half an hour to walk up there [ ] [ 35T: quick the time table 36A: well can’t we 37A: get that bus that goes from the Strand?= 38C: =yeh that’s what I’m saying but they only run every half an hour and I [ ] 39A: oh I see 40C: shall have to look up the times= → 41A: =right then sorry I’ve um [ 42C: oh that’s alright (laughs) [ 43A: messed up all the plans (.)no(.) I 44A: thought it was a joke when you said have you got the car [ 45T: ME? JOKE? 46C: yes I know(.) I knew you didn’t have YOUR car here(.) I thought you 47C: had your father’s here= 48A: =no this is er [ 49C: I presume it is your father’s car and not your mother’s [ 50A: no it’s my father’s car 51A: but my mother did bring me here because my father didn’t think he 52A: could stand going to see my gran (.) no if we go round there before 53A: they’re going on holiday she tends to get a bit er over emotional like (Data: 89.4) EXAMPLE 12 14C: 15A:

Gender and politeness

233

In other interactions between these three friends, it is clear that there is a great deal of give and take in the interactions: T, the male, is far quieter and intervenes less frequently than the women, partly because the women were at school together and they are more clearly friends with each other (although all of them have known each other since they were at school: they were in their thirties at the time of the recording). The misunderstanding about whether A has a car with her is something which is circled around in this conversation, in lines 8–14, where C first refers to jumping (out of a car) to find out from the bus depot when she has to work, and A in line 9 tries to check that this is C’s car which is being referred to: 7C: → 8C:

I’m going to phone them up in a minute or when we go out we can drive down and I’ll jump out [ → 9A: yeh is your car working?= 10C: =it’s it’s away at the moment = 11A =um= 12T: =at the hospital [ 13A: so how can we drive round there?= 14C: =I thought you said you had your car

As in many other misunderstandings in their conversations, conflicts are resolved through the use of humour and irony, with each of the participants trying to lessen the impact of the conflict, by using direct and forceful language, which might in Brown and Levinson’s model be categorised as FTAs, as well as more conventional apologies and negative politeness. Thus, when A apologises for not having a car with her and thus for having ‘messed up all the plans’ for going out, as they now have to take the bus, C states that she in fact knew that A did not have her car, contradicting her earlier statements that she had presumed she did have a car with her, a conventional negative politeness strategy, whereas in 45, T repeats A’s words in 44 ironically, presumably to lessen the tension of the situation. 41A: 42C: 43A: 44A:

=right then sorry I’ve um [ oh that’s alright (laughs) [ messed up all the plans (.)no(.) I thought it was a joke when you said have you got the car [

234 → 45T: → 46C: 47C:

Gender and Politeness ME? JOKE? yes I know(.) I knew you didn’t have YOUR car here(.) I thought you had your father’s here=

Focusing only on the way that A explicitly apologises, which conventional politeness theories such as Holmes do, would not allow us to focus on the way that in this interaction questions of the sex of the participants is not particularly salient. Analysing only the individual’s speech in relation to this apology also would not take into account the way that positive politeness or face work is being mutually accomplished, with each of the interactants contributing to the resolution of the misunderstanding with none of their faces being particularly threatened. The functioning of their particular community of practice is contributed to partly through the conflicts which go on in their group, that is, they can present themselves as a group of friends who get on well together because they can resolve conflicts jointly, not allowing difficulties and misunderstandings to threaten anyone’s face. For example, when A states in line 44, ‘I thought it was a joke when you said have you got the car’, T responds with irony: ‘ME (.)JOKE?’ drawing attention to the fact that, in fact, the group functions largely though the resolving of conflict through irony, banter, mockery, and joking. Even A’s apology in line 41, which seems like a straightforward apology, has the ring of irony about it, especially given the use of hesitation in the middle of the utterance. Thus, within this group it is difficult to make assumptions about the linguistic performance of the individuals on the basis of sex alone, since their production of their identities and roles within this particular community of practice is not determined solely by the gendering of their individual selves. This type of analysis is therefore trying to tease out where gender is a salient feature of the way that individuals present themselves within a group, and the way that they may draw on resources of polite behaviour and also impolite behaviour, ironically, in order to resolve conflicts. Conclusions In this chapter, I have tried to question the theoretical and analytical work which has been undertaken on the subject of women’s and men’s use of politeness. Contrary to Holmes’ and Brown’s work, which asserts a global difference between men’s and women’s use

Gender and politeness

235

of politeness, however mitigated some of their generalisations are, I should like to assert that gender ought not to be seen as a factor which determines the production or interpretation of speech in any simple way. That is not to say that gender is not important, as hypothesised stereotypes of feminine and masculine behaviour obviously play a role in the production of what participants see as appropriate or inappropriate speech. However, decisions about what is appropriate or not are decided upon strategically within the parameters of the community of practice and within the course of the interaction rather than being decided upon by each individual once and for all. The extracts I have analysed in this chapter show the difficulty of asserting that women and men speak in different ways, or of asserting that women behave in powerless ways, because it is clear that some participants use seemingly feminine strategies strategically to achieve their ends rather than to assert their subordination. They also collaborate with others to achieve certain ends within the group, and thus an analysis of their individual utterances alone would not enable an examination of the way that the group functions and the interaction takes shape. Because gender and other factors impact upon the context and because gender is indeed something which participants perform and interpret in the context of hypothesised gendered stereotypes within a community of practice, it is essential to analyse gender at both the local and the structural level, especially in its relation to the production and interpretation of politeness. Notes 1. Women are often viewed in certain contexts as a civilising force; for example, in the white colonisation of Australia, women were regarded as either whores or as ‘God’s police’, that is, as the guardians and enforcers of civilised norms (Robinson, 1988). This policing role is contrasted to the stereotypical Australian masculine role, which is associated with the outback and implicitly freedom from constraints (Schaffer, 1989). 2. These stereotypes of feminine behaviour, whilst being generated from middle-class women’s behaviour are often extended to working-class women’s behaviour as I mentioned in ch. 4 to distinguish between ‘respectable’ and ‘rough’ working-class women, both by others and the women themselves. 3. It might be argued that etiquette books are generally directed to members of the working classes who aspire to be considered as middle class, rather

236

4.

5.

6.

7.

8.

9. 10. 11.

Gender and Politeness than at the middle classes themselves, who might be assumed to know the ‘correct’ behaviour as part of their ‘cultural capital’ (Bourdieu, 1991). Distinctions between U and non-U behaviour (for example, whether to use the terms tea, dinner, or supper; sitting room or lounge; napkin or serviette, and so on) which were popular in Britain from the 1940s onwards still seem to have some resonance, although for many these distinctions seem less clear than they were. Feminist linguists and activists have often criticised this stereotypical association of women with politeness and have urged women to speak more assertively. However, strong women speakers, as I argued in ch. 4, may often be considered to be acting outside their allotted role, particularly by those who have a strong sense of ‘women’s place’ or the importance and value of femininity. Assertiveness may be judged to be aggression which is evaluated negatively for women. Anne Robinson, a British TV presenter who hosts the game show ‘The Weakest Link’ has been judged to have a particularly ‘cruel and po-faced’ persona (Braid, 2001). She was rated the rudest woman on TV, and journalists have termed her a ‘bitch’ and the ‘Miss Whiplash of prime-time TV’ (Braid, 2000). This type of behaviour is also extended from middle-class men to other men, and from middle-class women to other women, but at a stereotypical level, for historical reasons, because of its origins in chivalry, it is often associated with the behaviour of men to women. It certainly causes most conflict between men and feminist women; (others might argue that it is most appreciated between older groups of men and women). There is certainly an assumption in much media reporting that it is working-class men who are most prone to incidents of ‘road-rage’ and ‘air-rage’. Reindl Scheuering defines courtesy in the following way: courtesy is ‘the name given to practices such as a man helping a woman into her coat, leading her through crowded places or into a restaurant, opening her doors – the same as mothers do for children. See it this way and you know how men look on women’ (Reindl-Scheuering, cited in Kramarae and Treichler, 1985). For a more complex view of women’s use of prestige forms, see James (1996) ‘Women, men and prestige forms: a critical review’. This in itself is not problematic but the generalising about women and men as a whole from this data is. Baxter, in a conference paper at the International Gender and Language Association conference at Lancaster, 2002, stated that although girls often find it more difficult to speak in public than boys initially, this is because of their perception of stereotypes about appropriate behaviour, rather than any intrinsic quality.

Gender and politeness

237

12. However, it should be noted that compliments from men to women in a work environment may be interpreted as drawing attention to gender difference when it is not, or should not be, salient. 13. She also suggests that men and women tend to apologise for different infringements, suggesting that women apologise for time offences more than men: ‘keeping someone waiting is impolite behaviour and women tend to avoid being impolite more than men do’ (Holmes, 1995: 168). She even goes on to suggest that men are more often late than women. She gives other reasons for apologies occurring more: ‘it is perhaps not surprising to find a predominance of apologies for accidental body contact in a group who are the main victims of sexual harassment’ (Holmes, 1995: 169). However, we might argue that in this respect it should be men who apologise more because they are the ones who are most likely to be accused of sexual harassment.

Conclusions

This analysis of the complex relation between gender and politeness has implications for future research, not simply in the rather narrow field of politeness research but also within linguistics in general, and feminist linguistics in particular. I have also been arguing throughout this book for a rapprochement between Conversational Analysis and Critical Discourse Analysis, so that analysis of conversation can incorporate both openly political analysis, which focuses on negotiations with power relations at a structural and a local level, and also close textual analysis, which considers the way that individuals orient to those power relations. Language and gender research Second-Wave feminist linguistics drew on essentialist models of gender, and in many ways this is a very seductive form of analysis because experimental data can be made to fit this binary model of gender difference. Analysing gender as a simple factor which influences or determines language production for all women makes research and experimental work simpler. However, it is clear that if we simply assume a homogeneous male and female population, we cannot make accurate assessments of the role that gender plays. If we assume that stereotypes exist in a reified form which people simply accept or reject, we cannot account for the force of those stereotypes in people’s language production and reception, and in their negotiation of particular linguistic styles and subject positions. We also cannot account for change and difference in perceptions of stereotyping. If we ignore the role of stereotype completely in the production and reception of speech, it is clear that we will be unable 238

Conclusions

239

to assess the way that people come to judgements about other people’s speech. If we analyse gender in isolation from other factors like race and class, education and sexual orientation, and if we analyse gender as an individual ‘possession’ rather than as a factor which influences and shapes context and setting, we shall not be able to analyse the workings of gender adequately. Gender, as I have argued throughout this book, is not something which is simply imposed on individuals but is something which individuals work on and make their own, however uncomfortable and difficult that process is. Gender also cannot be simply correlated with the use of particular linguistic forms or strategies; as Ochs argues: ‘the relation between language and gender is mediated and constituted through a web of socially organised pragmatic meanings. Knowledge of how language relates to gender is not a catalogue of correlations between particular linguistic forms and sex of speakers, referents, addressees and the like. Rather, such knowledge entails tacit understanding of (1) how particular linguistic forms can be used to perform particular pragmatic work (such as conveying stance and social action) and (2) norms, preferences and expectations regarding the distribution of this work vis a vis particular social identities of speakers, referents, addressees’ (Ochs, cited by Brown, 1993: 139). Thus, the association of women with the use of tag-questions or with minimal responses, for example, is one which operates only at the level of stereotype, but this stereotype may have effects on the way interactants see themselves and their role within the community of practice. Third-Wave feminist linguistics is concerned to develop this type of anti-essentialist analysis of gender and language, whilst being keenly aware of the force of institutional and societal pressures on women and their resistances to those pressures. This has led to a move away from viewing women as victims to seeing the way that meanings and power relations are co-constructed in context. It has also led to a focus on the performative nature of gendered identity – the way that identity is worked out within the context of the constraints of communities of practice. This heterogeneous view of women means that factors of race and class cannot simply be considered in addition to gender (first we shall analyse the role of gender, and then race and class) but must be seen as instantiated and worked through at the same time as gender by women and

240

Gender and Politeness

men (Mills, forthcoming, b). Power must also be analysed in a more complex way, since not all women or Black and working-class people are powerless; using a more Foucauldian model of power, we can see that power is something which is worked out in our relations with others. Although institutional power and the power of the state does operate on us as individuals to map out for us our institutional status, we nevertheless achieve a local or interactional status through the way that we function as interactants in a range of different communities of practice. Thus, I am not advocating that we simply turn to the text and analyse the way that power relations are attended to by participants, although I feel that is significant; it is necessary as well to analyse the way that the resources available to participants are called upon in order to construct positions in the hierarchy of the group, as Diamond has suggested (Diamond, 1996). However, in addition to this, we also need to address the way that individuals do not only accrue interactional power for themselves through the conventional ‘masculine’ strategies of having a topic adopted by the group, reformulating utterances, and so on, but also how interactional power may be accrued by drawing on more ‘feminine’ strategies, such as resolving problems, ensuring that everyone has their speaking rights, and so on. We must ensure that in our move away from a Second-Wave feminist concern with the analysis of male and female speech patterns, we do not then go on to analyse masculinity and femininity, but implicitly assume that masculinity is necessarily more powerful. Different communities of practice will have different takes on what is counted as powerful. In thinking about the relationship between gender and power, we need to move beyond Second-Wave feminist concepts such as patriarchy, which seem to suggest that women are universally oppressed and that all men benefit from their oppression (Coward, 1983). Nevertheless, it is necessary still to be aware of the degree to which women are disadvantaged in relation to men in material terms: in terms of the salaries they earn; their positions in hierarchies; their representation in Parliament; the amount of housework and childcare they are expected to do; the degree that they are at risk of sexual assault and violence; and so on. Whilst within Third-Wave feminism, it is not now possible to say that all women are oppressed in similar ways and to the same degree, it does seem to be possible to argue that women are still systematically discriminated against,

Conclusions

241

and that this discrimination occurs at both a structural level (institutions and the state) and at a local level (relationships and families). Thus, Third-Wave feminism, whilst aware of the differences between groups of women, nevertheless stresses that women as a whole are subject to discrimination, but also emphasises women’s resistance to oppression. Therefore, whilst the notion of ‘women’ has been destabilised to a certain extent and is difficult to use except with provisos, it is still a concept which it is important to retain in order to be able to describe the systematic nature of the discrimination that many women experience (Modelski, 1991). As I mentioned above, current thinking on stereotyping is not adequate to account for the different perspectives individuals take. Even the more fragmented and conflictual notion of stereotyping developed by Bhabha within post-colonial theory, which allows us to see the way that stereotypes involve desire as well as negative Othering, nevertheless suggests that these stereotypes of gender and race are fixed and available to all in the same form, whereas in fact they are hypothesised and differently evaluated (Bhabha, 1994, 1990). The sheer repetition of versions of these stereotypes leads to a certain stability and institutionalised status. Some people who have invested in fashioning their identity in relation to what they think is stereotypical femininity will value traits associated with femininity, whereas those women who see those traits as confining or retrograde will construct an identity for themselves from different models, for example drawing on modes of self identity such as feminism, masculinity, or alternative politics (Holland, 2002). Those women and men who do affiliate or value stereotypical femininity will also have slightly different positions on politeness and what language they assume to be appropriate to men and to women. Thus, research in language and gender must focus on the local workings-out of gendered identities for interactants as well as analysing the larger processes whereby certain resources are available for participants. It might be thought that this anti-essentialist Third-Wave feminist focus on the workings-out of power will not permit any generalisations about gender to be made. However, it is clear that it is possible in the discussion of gender and politeness to assert that there are tendencies for certain classes of women within particular contexts to have available to them the resources of certain types of politeness behaviour. Their production of themselves and their relations

242

Gender and Politeness

to others through the use of politeness may be viewed by others critically or affirmatively, and thus this association of middle-class, white women with polite behaviour may be affirmed or contested at the level of stereotype. This association of middle-class women with politeness has major impacts not just on that group of women therefore but also on other women and men in other social classes and ethnic groups, and women and men in those other groups construct their own identities out of the resources of politeness which they consider to be available to them or which they consider to be off-limits. Linguistic analysis and politeness research As I have argued throughout this book, the current stand-off between Conversation Analysis and Critical Discourse Analysis is not theoretically productive. It is essential that each side of the theoretical divide begins to listen more carefully to the other, and to see the necessity of synthesising elements from each approach. This process is already in train, since many of the theorists working in these two fields are engaging in critiques of their disciplines (see, for example, Burman and Parker, 1993; Meyer, 2001; Wodak and Meyer, 2001). For many theorists, both in CA and CDA, it is essential to focus on the way that power relations affect those who suffer from inequality; as Van Dijk argues: ‘instead of focusing on purely academic or theoretical problems, [CDA] starts from prevailing social problems, and thereby chooses the perspective of those who suffer most, and critically analyses those in power, those who are responsible, and those who have the means and the opportunity to solve such problems’ (Van Dijk, cited in Wodak, 2001: 1). This is a perspective which I thoroughly endorse, yet at the same time it is important to be critical of this view of the workings of power. Here, the analyst is acting for those who suffer most, in order to bring about change, and power relations are characterised as simply a binary opposition between powerful and powerless. Political analysis needs to be more nuanced: as Thornborrow has argued, CDA analysts such as Fairclough ‘tend to see power as already accruing to some participants and not to others, and this power is determined by their institutional role and their socio-economic status, gender or ethnic identity’ (Thornborrow, 2002: 7). Burman and Parker argue

Conclusions

243

that we must be critical of any notion of linguistic analysis as simply empowerment, since the term itself masks ethically dubious bids for power (Burman and Parker, 1993). It is possible to act politically without erasing the agency of those in whose interests one sees one’s self as acting. Furthermore, as I have shown throughout this book, even those who have very little institutional power in fact may be able to resist those in positions of high status covertly or overtly. Whilst analysts must do what they can to bring about change, especially in the areas of racism and sexism, they nevertheless need to be aware of the agency of those ‘who suffer most’ and not characterise them as powerless. Many CDA linguists have been led to focus on the use of powerful language, such as racist and sexist discourse, and thus have tended to analyse individual linguistic items which are seen to be correlated with power and oppression. Instead of thinking that power is expressed in talk through the use of certain strategies or forms, we should think about ‘more or less powerful ways of talking’ within particular contexts: ‘it is not the linguistic form as such that is powerful or otherwise, rather it is more a question of who uses it and to what purpose that matters. In other words it is the use of language in context that will determine the function and the effects of an utterance and relations of power between speakers may well be central to that contextualised function’ (Thornborrow, 2002: 8). We need to see interactional power less as a variable that is achieved but as an ongoing assessment by participants in communities of practice. Some CA linguists have tried to move away from the notion of the variable impacting on participants in talk, and have therefore turned to the orientation of participants to certain issues such as power or gender. However, this has sometimes led to them ignoring political inequalities completely in this focus on the text itself. This textual focus can lead to even the most politically informed CA interpretation functioning as a simple empiricist analysis, which ultimately reinforces the status quo. CA theorists need to see that the way that groups orient to, and construct, the parameters of gender, race, and class has an impact on the limits within which participants feel that they can construct their notions of their identity. Interactants negotiate with stereotypes, with what they assume are societal values, and with the constraints of the community of practice in thinking about what it is possible for women and men to do and say. Thus,

244

Gender and Politeness

analysis of power within CDA and CA needs to be focused both on the local workings-out of power and also on the wider institutional level (and the relation between these levels) in order to move away from ‘the conception of discourses as if they were “tectonic plates” whose clashes constitute subjectivity’, since this ‘can present so distributed a notion of power that there is no room for agency, thus lapsing into mechanistic explanation’ (Burman and Parker, 1993: 163). Both CA and CDA have developed different analytic procedures. However, both of these sets of procedures are based on the assumption that conversation is relatively easy to analyse and that we as analysts can say what is going on in the conversation. As Burman and Parker argue, discourses do not simply emerge when we analyse an interaction closely, but rather ‘emerge as much through our work of reading as from the text’ (Burman and Parker, 1993: 156). This focus on the best possible reading of the interaction may lead us to ‘close the text to alternative readings’ (Burman and Parker, 1993: 156). We must instead, as analysts, be more modest in terms of what we can say about an interaction and, drawing on the work of integrationalist linguistics, see the transcribed text as one element in the process of making sense of discourse, and our own interpretation as a justified analysis of the text, but nevertheless still only one of many other interpretations.1 What I have been calling for is a greater consideration of the possibility of multiple interpretations for utterances and stretches of interaction, so that rather than assuming that each participant knows exactly what is going on in an interaction, we can allow that perhaps there is ambiguity in all interaction, and that certainty over intentions and interpretation is precarious and is achieved only fleetingly. Whilst it may be difficult to analyse exchanges using this approach to analysis, it may be closer to the way that participants themselves make sense of utterances. Politeness research cannot simply continue to use the Brown and Levinson model for analysis, or to modify Brown and Levinson’s work without questioning their underlying premisses about language and about politeness. So many theorists have criticised Brown and Levinson’s model but have only modified aspects of the model without fundamentally changing anything substantive about it. I would argue that it is necessary to challenge the sense that politeness can be simply described using models of language which only

Conclusions

245

analyse the speaker’s assumed intentions; which assume that politeness can be easily identified by the analyst; which do not analyse the role of assessment or judgement by speakers and hearers; and which do not have a complex model of the way that factors such as race, class, and gender have an impact on language production and interpretation. Because of the deficiencies of Brown and Levinson’s model in these respects, politeness research needs to move in significantly different directions to a consideration of context and social pressures, and other factors which lead speakers and hearers to judge that politeness or impoliteness is necessary or appropriate in particular contexts (Cross-Cultural Linguistic Politeness Research Group, 2002). Politeness cannot be seen as simply ‘nice’ or ‘deferent’ behaviour towards others, but should be seen as a wide-ranging set of behaviours which individuals view differently depending on the context and interactants. This set of behaviours and linguistic resources is one over which there are conflicts from moment to moment within communities of practice. Some members of groups may consider politeness to be a ‘good thing’ in general, but not something that they are committed to performing in terms of their own sense of themselves in that particular context at that particular moment. Because of the stereotypical association of middle-class, white women with politeness behaviour, it has to be seen that politeness cannot be analysed as if everyone had equal access to or equal investment in this type of behaviour. I am not suggesting that we entirely dispense with Brown and Levinson’s model, as I hope I have shown throughout this book that some elements of their work are extremely insightful in terms of being able to trace the sites where politeness behaviour may be taking place. Their model may also be useful in mapping out some of the mechanics of politeness and impoliteness processing. However, their model is not complex enough to deal with the way that politeness norms are negotiated by individuals in communities of practice, and the way that there are misunderstandings and breakdowns of communication over whether someone has been polite or impolite. Because of the complexities of politeness, only participants in the interaction can really know whether they consider something polite or not; however, I am not arguing that therefore it is impossible to analyse politeness at all. It is clearly a resource which interactants use to structure their relations with others, and they are able to be

246

Gender and Politeness

self-reflexive about both their own and others’ uses of politeness and impoliteness. Politeness can, therefore, be seen as a set of resources which is similar to money in that it is a way of structuring relations with others, but very often interactants are dealing in different currencies and different standards of exchange. I therefore advocate that linguistic analysis turns to the analysis of longer stretches of speech, and roots that analysis within communities of practice rather than focusing on individual participants. Consultation with participants is important to try to discover what they think is going on in the conversation, not so that this is considered the ‘truth’ about what was happening but so that the analyst can track down whether the participants’ hypothesised sense of appropriateness or stereotype is being brought into play. It is this complex to-and-fro movement between analysis of stereotype and appropriacy norms and the data itself over longer stretches of speech which might help us to move away from the use of individualistic models of interactants divorced from their social context. Note 1. Burman and Parker (1993: 168) argue that this relativist concern with seeing our own analysis as one among many may lead to casting our work as ‘fictive’, thereby making it difficult for us to make material interventions with our work. However, it is possible to justify one’s own findings and be aware of alternative readings for the interaction, some of them equally justified and some of them less so.

Bibliography

Adams, K. (1992) ‘Accruing power on debate floors’, in Hall. K., Bucholtz, M., and Moonwomon, B. (eds.) Locating Power, Berkeley, University of California, pp. 1–11. Afshar, H., and Maynard, M. (eds.) (1994) The Dynamics of ‘Race’ and Gender: Some Feminist Interventions, London, Taylor and Francis. Althusser, L. (1984) Essays on Ideology, London, Verso. Antaki, C. (1998) ‘Identity ascriptions in their time and place: “Fagin” and the “terminally dim”’, in Antaki, C., and Widdicombe, S. (eds.) Identities in Talk, London, Sage, pp. 71–86. Antaki, C., and Widdicombe, S. (1998) ‘Identity as an achievement and as a tool’, in Antaki, C., and Widdicombe, S. (eds.) Identities in Talk, London, Sage, pp. 1–14. Ariel, M., and Giora, R. (1992) ‘Gender versus group relation analysis of impositive speech acts’, in Hall, K. Bucholtz, M., and Moonwomon, B. (eds.) Locating Power, Berkeley, University of California, pp. 11–22. Aronson, K. (1991) ‘Social interaction and the recycling of legal evidence’, in Coupland, N., Giles, H., and Wiemann, J. (eds.) ‘Miscommunication’ and Problematic Talk, London, Sage. Atkinson, J. (1984) ‘Public speaking and audience responses: some techniques for inviting applause’, in Atkinson, J., and Heritage, J. (eds.) Structures of Social Action, Cambridge, Cambridge University Press. Atkinson, J., and Heritage, J. (1999) ‘Jefferson’s transcript notation’, in Jaworski, A., and Coupland, N. (eds.) The Discourse Reader, London, Routledge, pp. 158–166. Atkinson, J., and Heritage, J. (eds.) (1984) Structures of Social Action, Cambridge, Cambridge University Press. Azim, F. (1993) The Colonial Rise of the Novel, London, Routledge. Baker, S. (forthcoming) ‘Accent and context’, unpublished Ph.D. thesis, Colchester, University of Essex. Banks, S., Ge, G., and Baker, J. (1991) ‘Intercultural encounters and miscommunication’, in Coupland, N., Giles, H., and Wiemann, J. (eds.) (1991) ‘Miscommunication’ and Problematic Talk, London, Sage. 247

248

Bibliography

Baran, D., and Syska, O. (2000) ‘Harsh words are women’s words: the emergence of a new female speech style in Poland’, conference paper to the International Gender and Language Association, California, Stanford University. Bargiela, F. (2002) ‘Face politeness and self: new insights for old concepts’, discussion paper, Linguistic Politeness Research Group. Bargiela, F., Boz, C., Hamza, A., Tevzadze, M., and Mills, S. (2001) ‘Involvement strategies and naming’, in English Studies: Working Papers on the Web, www.shu.ac.uk/wpw Barthes, R. (1982) Barthes: Selected Writings, ed. S. Sontag, London, Fontana. Basso, E. (1992) ‘Contextualization in Kalapalo narratives’, in Duranti, A., and Goodwin, C. (eds.) (1992) Rethinking Context: Language as an Interactive Phenomenon, Cambridge, Cambridge University Press, pp. 253–271. Baxter, J. (2002) ‘Feminist post-structuralism and discourse analysis’, paper given to the International Gender and Language Association Conference, Lancaster. Beaken, M. (1996) The Making of Language, Edinburgh, Edinburgh University Press. Beebe, L. (1995) ‘Polite fictions: instrumental rudeness as pragmatic competence’, in Georgetown University Roundtable on Language and Linguistics, pp. 154–168. Bell, A. (1991) ‘Hot air: miscommunication and the climate change issue’, in Coupland, N., Giles, H., and Wiemann, J. (eds.) ‘Miscommunication’ and Problematic Talk, London, Sage. Bell, D., Binnie, J., Cream, J., and Valentine, G. (1994) ‘All hyped up and no place to go’, in Gender Place and Culture, 1/1, pp. 31–49. Bergvall, V., Bing, J., and Freed, A. (eds.) (1996) Rethinking Language and Gender Research: Theory and Practice, London, Longman. Berk-Seligson, S. (1988) ‘The impact of politeness on witness testimony: the influence of the court interpreter’, Multilingua, 7/4, pp. 411–439. Bhabha, H. (1994) ‘The other question: stereotype, discrimination and the discourse of colonialism’, in Bhabha, H. (ed.) The Location of Culture, London, Routledge. Bhabha, H. (ed.) (1990) Nation and Narration, London, Routledge. Bing, J., and Bergvall, J. (1996) ‘The question of questions: beyond binary thinking’, in Bergvall, V., Bing, J., and Freed, A. (eds.) Rethinking Language and Gender Research: Theory and Practice, London, Longman, pp. 1–30. Birkner, K., and Kirn, F. (2000) ‘Impression management in East and West German job interviews’, in Spencer-Oatey, H. (ed.) Culturally Speaking: Managing Rapport through Talk Across Cultures, London, Continuum, pp. 255–271. Blum-Kulka, S. (1992) ‘The metapragmatics of politeness in Israeli society’, in Watts, R., Ide, S., and Ehlich, K. (eds.) Politeness in Language:

Bibliography

249

Studies in Its History, Theory and Practice, Berlin and New York, Mouton de Gruyter, pp. 255–280. Bolton M., and Bolton, J. (n.d.) The Complete Book of Etiquette, Slough, Foulsham. Bourdieu, P. (1991) Language and Symbolic Power, Cambridge, Polity Press. Boz, C. (in progress) ‘Cultural difference and academic writing’, Ph.D. thesis, Sheffield, Sheffield Hallam University. Bradac, J., and Mulac, A. (1984) ‘A molecular view of powerful and powerless speech styles: attributional consequences of specific language features and communicator intentions’, Communication Monographs, 51, pp. 307–319. Braid, M. (2001) ‘Cruella of prime time’, Independent, 10 March, p. 5. Brown, P. (1993) ‘Gender, politeness and confrontation in Tenejapa’, in Tannen, D. (ed.) Gender and Conversational Interaction, Oxford, Oxford University Press. (1980) ‘How and why are women more polite: some evidence from a Mayan community’, in McConnell-Ginet, S., Borker, R., and Furman, N. (eds.) Women and Language in Literature and Society, New York, Praeger, pp. 111–136. Brown, P., and Levinson, S. (1987) Politeness: Some Universals in Language Usage, Cambridge, Cambridge University Press. (1978) ‘Universals in language usage: politeness phenomena’, in Goody, E. (ed.) Questions and Politeness: Strategies in Social Interaction, Cambridge, Cambridge University Press, pp. 56–311. Bucholtz, M. (1999a) ‘Bad examples: transgression and progress in language and gender studies’, in Bucholtz, M., Liang, A., and Sutton, L. (eds.) (1999) Reinventing Identities: The Gendered Self in Discourse, New York and Oxford, Oxford University Press. (1999b) ‘Why be normal? Language and identity practices in a community of nerd girls’, in Language in Society, 28/2, pp. 203–225. (1996) ‘Black feminist theory and African American women’s linguistic practice’, in Bergvall, V., Bing, J., and Freed, A. (eds.) Rethinking Language and Gender Research: Theory and Practice, London, Longman, pp. 267–290. Bucholtz, M., Liang, A., and Sutton, L. (eds.) (1999) Reinventing Identities: The Gendered Self in Discourse, New York and Oxford, Oxford University Press. Burman, E., and Parker, I. (1993) ‘Against discursive imperialism, empiricism and constructionism: thirty-two problems with discourse analysis’, in Burman, E., and Parker, I. (eds.) Discourse Analytic Research: Repertoires and Readings of Texts in Action, London and New York, Routledge. Butler, J. (1997) Excitable Speech: A Politics of the Performative, London, Routledge.

250

Bibliography

(1993) Bodies that Matter: On the Discursive Limits of Sex, London, Routledge. (1990) Gender Trouble: Feminism and the Subversion of Identity, London, Routledge. Cameron, D. (2000) Good to Talk? Living and Working in a Communication Culture, London, Sage. (1998a) ‘“Is there any ketchup, Vera?”: gender, power and pragmatics’, in Discourse and Society, 9/4, pp. 435–455. (1997) ‘Performing gender identity: young men’s talk and the construction of heterosexual masculinity’, in Johnson, S., and Meinhoff, U. (eds.) Language and Masculinity, Oxford, Blackwell, pp. 86–107. (1995) Verbal Hygiene, London, Routledge. Cameron, D. (ed.) (1998b) The Feminist Critique of Language: A Reader, 2nd edn, London, Routledge. Cameron, D., McAlinden, F., and O’Leary, K. (1988) ‘Lakoff in context: the social and linguistic functions of tag questions’, in Coates, J., and Cameron, D. (eds.) Women in their Speech Communities, Harlow, Longman, pp. 13–26. Chan. G. (1992) ‘Gender, roles and power in dyadic conversation’, in Hall, K., Bucholtz, M., and Moonwomon, B. (eds.) Locating Power, Berkeley, University of California, pp. 57–67. Chomsky, N. (1988) Language and Problems of Knowledge, Cambridge, Mass., MIT Press. Choulariaki, L., and Fairclough, N. (1999) Discourse in Late Modernity, Edinburgh, Edinburgh University Press. Christie, C. (2002) ‘Politeness and the linguistic construction of gender in Parliament: an analysis of transgressions and apology behaviour’, in Linguistic Politeness Research Group (ed.) Working Papers on the Web, www.shu.ac.uk/wpw (2000) Gender and Language: Towards a Feminist Pragmatics, Edinburgh, Edinburgh University Press. Cicourel, A. (1992) ‘The interpenetration of communicative contexts: examples from medical encounters’, in Duranti, A., and Goodwin, C. (eds.) (1992) Rethinking Context: Language as an Interactive Phenomenon, Cambridge, Cambridge University Press, pp. 291–311. Clark, K. (1998) ‘The linguistics of blame: representations of women in The Sun’s reporting of crimes of sexual violence’, in Cameron, D. (ed.) The Feminist Critique of Language: A Reader, 2nd edn, London, Routledge, pp. 183–197. Coates, J. (1999) ‘Changing femininities: the talk of teenage girls’, in Bucholtz, M., Liang, A., and Sutton, L. (eds.) (1999) Reinventing Identities: The Gendered Self in Discourse, New York and Oxford, Oxford University Press. (1997) ‘One-at-a-time: the organisation of men’s talk’, in Johnson, J., and Meinhof, U. (eds.) Language and Masculinity, Oxford, Blackwell, pp. 107–130.

Bibliography

251

(1996) Woman Talk, Oxford, Blackwell. (1988) ‘Gossip revisited’, in Coates, J., and Cameron, D. (eds.) Women in Their Speech Communities, London, Longman. Coates, J., and Cameron, D. (eds.) (1988) Women in Their Speech Communities, London, Longman. Cohen, R. (1987) ‘Problems of intercultural communication in EgyptianAmerican diplomatic relations’, International Journal of Intercultural Relations, 11, pp. 29–47. Coleman, L., and de Paulo, B. (1991) ‘Uncovering the human spirit: moving beyond disability and “missed” communication’, in Coupland, N., Giles, H., and Wiemann, J. (eds.) ‘Miscommunication’ and Problematic Talk, London, Sage. Conley, J., and O’Barr, W. (1990) ‘Rule versus relationships in small claims disputes’, in Grimshaw, A. (ed.) Conflict talk: Sociolinguistic Investigations of Arguments in Conversations, Cambridge, Cambridge University Press. Corsaro, W., and Rizzo, T. (1990) ‘Disputes in the peer culture of American and Italian nursery schoolchildren’, in Grimshaw, A. (ed.) (1990) Conflict Talk: Sociolinguistic Investigations of Arguments in Conversations, Cambridge, Cambridge University Press. Coulmas, F. (1992) ‘Linguistic etiquette in Japanese society’, in Watts, R., Ide, S., and Ehlich, K. (eds.) Politeness in Language: Studies in Its History, Theory and Practice, Berlin and New York, Mouton de Gruyter, pp. 298–323. Coupland, N., Giles, H., and Wiemann, J. (1991) ‘Talk as “problem” and communication as “miscommunication”: an integrative analysis’, in Coupland, N., Giles, H., and Wiemann, J. (eds.) ‘Miscommunication’ and Problematic Talk, London, Sage. Coward, R. (1983) Patriarchal Precedents: Sexuality and Social Relations, London, Routledge and Kegan Paul. Crawford, M. (1995) Talking Difference: On Gender and Language, London, Sage. Crawford, M., and Chaffin, R. (1986) ‘The reader’s construction of meaning; cognitive research of gender and comprehension’, in Flynn, E., and Schweickart, P. (eds.) Gender and Reading: Essays on Readers Texts and Contexts, Baltimore, Johns Hopkins University Press. Crompton, R., and Mann, M. (eds.) (1986) Gender and Stratification, London, Polity Press. Cross-Cultural Linguistic Politeness Research Group (ed.) (2002) Politeness and Context, Working Papers on the Web, 3, www.shu.ac.uk/wpw Culpeper, J. (2001) Language and Characterisation: People in Plays and Other Texts, Harlow, Longman. (1996) ‘Towards an anatomy of impoliteness’, in Journal of Pragmatics, 25, pp. 349–367. Daley, J. (2001) ‘Rude Britannia, a nation transformed’, in Courtesy Call, Newcastle, Polite Society, p. 3.

252

Bibliography

Diamond, J. (1996) Status and Power in Verbal interaction: A Study of Discourse in a Close-knit Social Network, Amsterdam/Philadelphia, John Benjamins. Davidoff, L., and Hall, C. (1995) ‘Gender and the middle class’, in Joyce, P. (ed.) Class, Oxford, Oxford University Press, pp. 302–316. Day, D. (1998) ‘Being ascribed and resisting membership of an ethnic group’, in Antaki, C., and Widdicombe, S. (eds.) Identities in Talk, London, Sage, pp. 151–169. Drummond, K. (1991) ‘Misunderstanding and its remedies: telephone miscommunication’, in Coupland, N., Giles, H., and Wiemann, J. (eds.) ‘Miscommunication’ and Problematic Talk, London, Sage. Duranti, A. (1992) ‘Language in context and language as context: the Samoan respect vocabulary’, in Duranti, A., and Goodwin, C. (eds.) (1992) Rethinking Context: Language as an Interactive Phenomenon, Cambridge, Cambridge University Press, pp. 77–101. Duranti, A., and Goodwin, C. (eds.) (1992) Rethinking Context: Language as an Interactive Phenomenon, Cambridge, Cambridge University Press. Eckert, P. (2000) Linguistic Variation as Social Practice, Oxford, Blackwell. (1989) ‘The whole woman: sex and gender differences in variation’, in Language Variation and Change, 1, pp. 245–267. Eckert, P., and McConnell-Ginet, S. (1999) ‘New generalisations and explanations in language and gender research’, in Language in Society, 28/2, pp. 185–203. (1998) ‘Communities of practice: where language, gender and power all live’, in Coates, J. (ed.) Language and Gender: A Reader, Oxford, Blackwell, pp. 484–494. Edley, N., and Wetherell, M. (1997) ‘Jockeying for position: the construction of masculine identities’, Discourse and Society, 8/2, pp. 203–217. Edwards, D. (1998) ‘The relevant thing about her: social identity categories in use’, in Antaki, C., and Widdicombe, S. (eds.) Identities in Talk, London, Sage, pp. 15–33. Edwards, V. (1988) ‘The speech of British Black women in Dudley, West Midlands’, in Coates, J., and Cameron, D. (eds.) Women in their Speech Communities, Harlow, Longman, pp. 33–51. Eelen, G. (2001) Critique of Politeness Theories, Manchester, St Jeromes Press. Ehlich, K. (1992) ‘On the historicity of politeness’, in Watts, R., Ide, S., and Ehlich, K. (eds.) Politeness in Language: Studies in its History, Theory and Practice, Berlin and New York, Mouton de Gruyter, pp. 71–108. Elsenberg, E., and Phillips, S. (1991) ‘Miscommunication in organisations’, in Coupland, N., Giles, H., and Wiemann, J. (eds.) ‘Miscommunication’ and Problematic Talk, London, Sage. Erlich, S. (1999) ‘ Communities of practice, gender and the representation of sexual assault’, in Language in Society, 28/2, pp. 239–257.

Bibliography

253

Erlich, S., and King, R. (1996) ‘Consensual sex or sexual harassment: negotiating meaning’, in Bergvall, V. et al. (eds.) Rethinking Gender and Language Research: Theory and Practice, London, Longman, pp. 153–173. Fairclough, N. (2000) New Labour, New Language, London, Routledge. (1995) Critical Discourse Analysis: The Critical Study of Language, Harlow, Longman. (1992) Discourse and Social Change, London, Polity Press. (1989) Language and Power, Harlow, Longman. Foley, W. (1997) Anthropological Linguistics, Oxford, Blackwell. Foucault, M. (1994/1972) The Archaeology of Knowledge, London, Tavistock. (1981) ‘The order of discourse’, in Young, R. (ed.) Untying the Text: A Poststructuralist Reader, London, Routledge and Kegan Paul. (1978) History of Sexuality: An Introduction, vol. i, Harmondsworth, Penguin. Frankenberg, R. (1993) White Women, Race Matters: The Social Construction of Whiteness, London, Routledge. Fraser, B., and Nolen, W. (1981) ‘The association of deference with linguistic form’, in International Journal of the Sociology of Language, 27, pp. 93–110. Freed, A. (1999) ‘Communities of practice and pregnant women: is there a connection?’, in Language in Society, 28/2, pp. 257–271. (1996) ‘Language and gender research in an experimental setting’, in Bergvall, V., Bing, J., and Freed, A. (eds.) Rethinking Language and Gender Research: Theory and Practice, Harlow, Longman, pp. 54–76. Fukushima, S. (2000) Requests and Culture: Politeness in British English and Japanese, Bern, Peter Lang. Fuss, D. (1989) Essentially Speaking: Feminism Nature and Difference, London, Routledge. Gaik, F. (1992) ‘Radio talk-show therapy and the pragmatics of possible worlds’, in Duranti, A., and Goodwin, C. (eds.) (1992) Rethinking Context: Language as an Interactive Phenomenon, Cambridge, Cambridge University Press, pp. 271–291. Gass, S., and Varonis, E. (1991) ‘Miscommunication in non-native speaker discourse’, in Coupland, N., Giles, H., and Wiemann, J. (eds.) ‘Miscommunication’ and Problematic Talk, London, Sage. Gavins, J. (2002) ‘Text Worlds and Absurdist Literature’, unpublished Ph.D. thesis, Sheffield, Sheffield Hallam University. Gilbert, S., and Gubar, S. (1988) The War of the Words, vol. i, New Haven, Yale University Press. Goffman, E. (1967) Interaction Ritual: Essays on Face to Face Behaviour, Harmondsworth, Penguin. (1999/1967) ‘On face-work: an analysis of ritual elements in social interaction’, in Jaworski, A., and Coupland, N. (eds.) The Discourse Reader, London, Routledge, pp. 306–321.

254

Bibliography

Goodwin, C. (1984) ‘Notes on story structure and the organisation of participation’, in Atkinson, J., and Heritage, J. (eds.) Structures of Social Action, Cambridge, Cambridge University Press. (1981) Conversational Organisation: Interaction Between Speakers and Hearers, New York and London, Academic Press. Goodwin, C., and Harness Goodwin, M. (1990) ‘Interstitial argument’, in Grimshaw, A. (ed.) Conflict Talk: Sociolinguistic Investigations of Arguments in Conversations, Cambridge, Cambridge University Press. (1992) ‘Assessments and the construction of context’, in Duranti, A., and Goodwin, C. (eds.) (1992) Rethinking Context: Language as an Interactive Phenomenon, Cambridge, Cambridge University Press, pp. 147–191. Goodwin, C. and Duranti, A. (1992) ‘Rethinking context: an introduction’, in Duranti, A., and Goodwin, C. (eds.) (1992) Rethinking Context: Language as an Interactive Phenomenon, Cambridge, Cambridge University Press, pp. 1–43. Grainger, K. (2002) ‘Politeness or impoliteness? Verbal play on the hospital ward’, in English Studies: Working Papers on the Web, http://www/shu.ac.uk/wpw.wpw.htm Greatbatch, D., and Dingwall, R. (1998) ‘Talk and identity in divorce mediation’, in Antaki, C., and Widdicombe, S. (eds.) Identities in Talk, London, Sage, pp. 121–132. Gregory, I. (2001a) ‘Nation on a short fuse’, in Courtesy Call: The Magazine of the Campaign for Courtesy (Summer) Newcastle, Polite Society, p. 1. (2001b) ‘First name “vulgarity”’, in Courtesy Call: The Magazine of the Campaign for Courtesy (May) Newcastle, Polite Society, p. 2. (2001c) ‘The new barbarians’, in Courtesy Call: The Magazine of the Campaign for Courtesy (May) Newcastle, Polite Society, p. 4. Grice, H. P. (1975) ‘Logic and conversation’, in Cole, P., and Morgan, J. (eds.) Syntax and Semantics, vol. iii, New York, Academic Press, pp. 41–58. Griffiths, V., and Ali Taha, A. (1936) Sudan Courtesy Customs, Khartoum, Sudan Government. Grimshaw, A. (1990) ‘Research on conflict talk’, in Grimshaw, A. (ed.) Conflict Talk: Sociolinguistic Investigations of Arguments in Conversations, Cambridge, Cambridge University Press. Grimshaw, A. (ed.) (1990a) Conflict Talk: Sociolinguistic Investigations of Arguments in Conversations, Cambridge, Cambridge University Press. Gudykunst, W. (2000) ‘Methodological issues in conducting theory-based cross-cultural research’, in Spencer-Oatey, H. (ed.) Culturally Speaking: Managing Rapport through Talk Across Cultures, London, Continuum, pp. 293–315.

Bibliography

255

Gumperz, J. (1982) ‘Contextualization and understanding’, in Duranti, A., and Gumperz, J. (eds.) (1982) Discourse Strategies, Cambridge, Cambridge University Press, pp. 229–253. Gunthner, S. (2000) ‘Argumentation and resulting problems in the negotiation of rapport in a German–Chinese conversation’, in Spencer-Oatey, H. (ed.) Culturally Speaking: Managing Rapport through Talk Across Cultures, London, Continuum, pp. 217–239. Hall, K., Bucholtz, M., and Moonwomon, B. (eds.) (1992) Locating Power: Proceedings of the Second Berkeley Women and Language Conference, vol. I, Berkeley, California Berkeley Women and Language Group, University of California. Hall, S. (1980) ‘Encoding/decoding’, in Hall, S., Hobson, A., Lowe, A., and Willis, P. (eds.) Culture, Media, Language, London, Hutchinson, pp. 128–138. Halberstam, J. (1998) Female Masculinity, London, Routledge. Hamza, A. (forthcoming) ‘A Cross-cultural Analysis of Politeness in NativeNon Native Interaction’, Ph.D. thesis, Sheffield, Sheffield Hallam University. Hanks, W. (1992) ‘The indexical ground of deictic reference’, in Duranti, A., and Goodwin, C. (eds.) (1992) Rethinking Context: Language as an Interactive Phenomenon, Cambridge, Cambridge University Press, pp. 43–77. Harness Goodwin, M. (2001) ‘Organising participation in cross-sex jump rope: situating gender differences within longitudinal studies of activities’, in Research on Language and Social Interaction, 34/1, pp. 75–106. (1998) ‘Games of stance: conflict and footing in hopscotch’, in Hoyle, S., and Adger, C. (eds.) Kids Talk: Strategic Language Use in Later Childhood, Oxford and New York, Oxford University Press, pp. 22–45. Harris, R. (1990) ‘Lars Porsena revisited’, in Ricks, C., and Michaels, L. (eds.) The State of the Language, London, Faber. Harris, S. (forthcoming 2003) ‘Politeness and power: making and responding to “requests” in institutional settings’. (2001) ‘Being politically impolite: extending politeness theory to adversarial political discourse’, in Discourse and Society, 12/4, pp. 451–472. Harris Bond, M., Zegarac, V., and Spencer-Oatey, H. (2000) ‘Culture as an explanatory variable: problems and possibilities’, in Spencer-Oatey, H. (ed.) Culturally Speaking: Managing Rapport through Talk Across Cultures, London, Continuum, pp. 47–76. Haugen, J. (2000) ‘Unladylike divas: the gender performances of female gangsta rappers’, conference paper to International Gender and Language Association, Stanford University, California. Haverkate, H. (1988) ‘Toward a typology of politeness strategies’, in Multilingua, 7/4, pp. 385–409.

256

Bibliography

Held, G. (1992) ‘Politeness in linguistic research’, in Watts, R., Ide, S., and Ehlich, K. (eds.) Politeness in Language: Studies in Its History, Theory and Practice, Berlin and New York, Mouton de Gruyter, pp. 131–155. Henley, N. (1995) ‘Ethnicity and gender issues in language’, in Landrine, H. (ed.) Bringing Cultural Diversity to Feminist Psychology: Theory, Research and Practice, Washington, American Psychological Association. Henley, N., and Kramarae, C. (1991) ‘Gender power and miscommunication’, in Coupland, N., Giles, H., and Wiemann, J. (eds.) ‘Miscommunication’ and Problematic Talk, London, Sage. Hewitt, R. (1997) ‘“Box-out” and “Taxing”’, in Johnson, S., and Meinhof, U. (eds.) Language and Masculinity, Oxford, Blackwell, pp. 27–47. Hird, B. (1996) ‘EAP: teaching Chinese learners to be impolite’, English in Australia, 115, March, pp. 29–37. Holland, S. (2002) ‘Not Just Girly: Alternative Women and Femininity’, Ph.D. thesis, Sheffield, Sheffield Hallam University. Holmes, J. (1995) Women, Men and Politeness, London, Longman. Holmes, J. and Meyerhoff, M. (1999) ‘The community of practice: theories and methodologies in language and gender research’, in Language in Society, 28/2, pp. 173–185. House, J. (2000) ‘Understanding misunderstanding: a pragmatic-discourse approach to analysing mismanaged rapport in talk across cultures’, in Spencer-Oatey, H. (ed.) Culturally Speaking: Managing Rapport through Talk Across Cultures, London, Continuum, pp. 145–164. Ide, S., Hill, B., Carnes, Y., Ogino, T., and Kawasaki, A. (1992) ‘The concept of politeness: an empirical study of American English and Japanese’, in Watts, R., Ide, S., and Ehlich, K. (eds.) Politeness in Language: Studies in its History, Theory and Practice, Berlin and New York, Mouton de Gruyter, pp. 281–297. Jackson, P. (1994) ‘Black male: advertising and the cultural politics of masculinity’, in Gender Place and Culture, 1/1, pp. 49–61. James, D. (1996) ‘Women, men and prestige speech forms: a critical review’, in Bergvall, V., Bing, J., and Freed, A. (eds.) (1996) Rethinking Language and Gender Research: Theory and Practice, London, Longman, pp. 98–125. Janney, R., and Arndt, H. (1992) ‘Intracultural tact versus intercultural tact’, in Watts, R., Ide, S., and Ehlich, K. (eds.) Politeness in Language: Studies in Its History, Theory and Practice, Berlin and New York, Mouton de Gruyter, pp. 21–42. Jarrett-Macauley, D. (ed.) (1996) Reconstructing Womanhood: Reconstructing Feminism: Writings on Black Women, London, Routledge. Jary, M. (1998) ‘Relevance theory and the communication of politeness’, in Journal of Pragmatics, 30, pp. 1–19.

Bibliography

257

Jefferson, G. (1984a) ‘On the organisation of laughter in talk about troubles’, in Atkinson, J., and Heritage, J. (eds.) Structures of Social Action, Cambridge, Cambridge University Press. (1984b) ‘In stepwise transition from talk about trouble to inappropriately next positioned matters’, in Atkinson, J., and Heritage, J. (eds.) Structures of Social Action, Cambridge, Cambridge University Press. Johnson, S., and Meinhof, U. (eds.) (1997) Language and Masculinity, Oxford, Blackwell. Johnstone, B., Ferrara, K., and Mattson Bean, J. (1992) ‘Gender, politeness and discourse management in same-sex and cross-sex opinion poll interviews’, in Journal of Pragmatics, 18, pp. 405–430. Joyce, P. (ed.) (1995) Class, Oxford, Oxford University Press. Jucker, A. (1988) ‘The relevance of politeness’, in Multilingua, 7/4, pp. 375–384. Kasper, G. (2000) ‘Data collection in pragmatics research’, in SpencerOatey, H. (ed.) Culturally Speaking: Managing Rapport through Talk Across Cultures, London, Continuum, pp. 316–341. Keinpointner, M. (1997) ‘Varieties of rudeness: types and functions of impolite utterances’, Functions of Language, 4/2, pp. 251–287. Kendall, G., and Wickham, G. (1999) Using Foucault‘s Methods, London, Sage. Kendon, A. (1992) ‘The negotiation of context in face-to-face interaction’, in Duranti, A., and Goodwin, C. (eds.) (1992) Rethinking Context: Language as an Interactive Phenomenon, Cambridge, Cambridge University Press, pp. 323–335. Kharraki, A. (2001) ‘Moroccan sex-based linguistic difference in bargaining’, Discourse and Society, 12/5, pp. 615–632. De Klerk, V. (1997) ‘The role of expletives in the construction of masculinity’, in Johnson, S., and Meinhof, U. (eds.) Language and Masculinity, Oxford, Blackwell, pp. 144–159. Kramarae, C., and Treichler, P. (eds.) (1985) A Feminist Dictionary, London, Pandora. Kristeva, J. (1984) Revolution in Poetic Language, New York, Columbia University Press. (1982) Powers of Horror: An Essay on Abjection, New York, Columbia University Press. Labov, W. (1990) ‘The intersection of sex and social class in the course of linguistic change’, in Language Variation and Change, 2, pp. 205–254. (1972) Language in the Inner City, Philadelphia, University of Pennsylvania Press. Lakoff, G. (1987) Women, Fire and Dangerous Things: What Categories Reveal about the Mind, Chicago, Chicago University Press. Lakoff, R. (2000) The Language Wars, Berkeley, University of California Press. (1975) Language and Woman’s Place, New York, Harper and Row.

258

Bibliography

Langford, P. (1989) A Polite and Commercial People: England 1727–1783, Oxford, Oxford University Press. Leap, W. (ed.) (1995) Beyond the Lavender Lexicon: Authenticity, Imagination and Appropriation in Lesbian and Gay Languages, Luxembourg, Gordon and Breach. Leech, G. (1983) Principles of Pragmatics, Longman, Harlow. Lewis, R., and Mills, S. (eds.) (forthcoming) Postcolonial Feminist Theory: A Reader, Edinburgh, Edinburgh University Press. Liladhar, J. (2001) ‘Making, Unmaking and Re-Making Femininity’, unpublished Ph.D. thesis, Sheffield, Sheffield Hallam University. Lindstrom, L. (1992) ‘Context contests: debatable truth statements on Tanna (Vanuatu)’, in Duranti, A., and Goodwin, C. (eds.) (1992) Rethinking Context: Language as an Interactive Phenomenon, Cambridge, Cambridge University Press, pp. 101–125. Livia, A., and Hall, K. (eds.) (1997) Queerly Phrased: Language Gender and Sexuality, New York and Oxford, Oxford University Press. Lycan, W. (1977) ‘Conversation, politeness and interruption’, Papers in Linguistics, 1–2, 10 Spring, pp. 23–53. Manke, M. (1997) Classroom Power Relations: Understanding Student Teacher Interaction, New Jersey and London, Lawrence Erlbaum Associates. Mao, L. (1994) ‘Beyond politeness theory: “face” revisited and renewed’, in Journal of Pragmatics, 21/5, pp. 451–486. Matsuda, M., Lawrence, C., Delgado, R., and Crenshaw, K. eds (1993) Words that Wound: Critical Race Theory, Boulder/San Francisco: Westview Press. Maynard, M., and Purvis, J. (ed.) (1995) (Hetero) sexual Politics, London, Taylor and Francis. McClintock, A. (1995) Imperial Leather: Race, Gender and Sexuality in the Imperial Contest, London, Routledge. McElhinny, B. (1998) ‘“I don’t smile much anymore”: affect, gender and the discourse of Pittsburgh Police Officers’, in Coates, J. (ed.) Language and Gender: A Reader, Oxford, Blackwell, pp. 309–327. McIlvenny, P. (2002) ‘CDA and CA’, paper to International Gender and Language Association conference, Lancaster University. McKinlay, A., and Dunnett, A. (1998) ‘How gun-owners accomplish being deadly average’, in Antaki, C., and Widdicombe, S. (eds.) Identities in Talk, London, Sage, pp. 34–51. McTear, M., and King, F. ‘Miscommunication in clinical contexts’, in Coupland, N., Giles, H., and Wiemann, J. (eds.) ‘Miscommunication’ and Problematic Talk, London, Sage. Mehan, H. (1990) ‘Oracular reasoning in a psychiatric examination’, in Grimshaw, A. (ed.) Conflict Talk: Sociolinguistic Investigations of Arguments in Conversations, Cambridge, Cambridge University Press.

Bibliography

259

Meyer, M. (2001) ‘Between theory, method and politics: positioning of the approaches of CDA’, in Wodak, R., and Meyer, M. (eds.) Critical Discourse Analysis, London, Sage. Meyerhoff, M. (1999) ‘Sorry in the Pacific: defining communities, defining practices’, in Language in Society, 28/2, pp. 225–239. Miller, L. (2000) ‘Negative assessments in Japanese-American workplace interaction’, in Spencer-Oatey, H. (ed.) Culturally Speaking: Managing Rapport through Talk Across Cultures, London, Continuum, pp. 240–254. Mills, S. (forthcoming,a) Michel Foucault, London, Routledge. (forthcoming,b) Third Wave Feminist Linguistics and the Analysis of Sexism, Cambridge, Cambridge University Press. (2002) ‘Class, gender and politeness’ paper to the Politeness and Power conference, Loughborough University. (2002) ‘Rethinking politeness, impoliteness and gender identity’, in Sunderland, J., and Litoselliti, L. (eds.) Discourse Analysis and Gender Identity, Amsterdam/Philadelphia, Benjamins. (1999) ‘Discourse competence: or how to theorise strong women speakers’, in Hendricks, C., and Oliver, K. (eds.) Language and Liberation: Feminism, Philosophy and Language, New York, State of New York University Press, pp. 81–99. (1998) ‘Post-feminist text analysis’, in Language and Literature, 7/3, pp. 235–254. (1997) Discourse, London, Routledge. (1996) ‘Powerful talk’, unpublished discussion paper, Loughborough, Loughborough University. (1995a) Feminist Stylistics, London, Routledge. (1995b) ‘Discontinuity and post-colonial discourse’, in Ariel: A Review of International English Literature, 26/3, pp. 73–89. (1994a) ‘Introduction’, in Mills, S. (ed.) Gendering the Reader, Hemel Hempstead, Harvester, pp. 1–23. (1994b) ‘Reading as/like a feminist’, in Mills, S. (ed.) Gendering the Reader, Hemel Hempstead, Harvester, pp. 25–47. Minh-ha, T. (1989) Woman Native Other: Writing Postcoloniality and Feminism, Bloomington, Indian University Press. Modleski, T. (1991) Feminism Without Women: Culture and Criticism in a ‘Postfeminist’ Age, London and New York, Routledge. Mohanty, C. (1984) ‘Under Western eyes: feminist scholarship and colonial discourse’, in Boundary 2, vol. iii, pp. 333–358. Montgomery, M. (1999) ‘Speaking sincerely: public reactions to the death of Diana’, Language and Literature, 8/1, pp. 5–33. Morgan, M. (1999) ‘No woman no cry: claiming African American women’s place’, in Bucholtz, M., Liang, A., and Sutton, L. (eds.) (1999) Reinventing Identities: The Gendered Self in Discourse, New York and Oxford, Oxford University Press.

260

Bibliography

Nichols, P. (1998) ‘Black women in the rural South: conservative and innovative’, in Coates, J. (ed.) Language and Gender: A Reader, Oxford, Blackwell, pp. 55–63. Nussbaum, J., and Coupland, N. (1991) ‘The reproduction of ageing and ageism in intergenerational talk’, in Coupland, N., Giles, H., and Wiemann, J. (eds.) ‘Miscommunication’ and Problematic Talk, London, Sage. O’Barr, W. and Atkins, B. (1980) ‘“Women’s language” or “powerless language”?’, in McConnell-Ginet, S., Borker, R., and Furman, N. (eds.) Women and Language in Literature and Society, New York, Praeger, pp. 93–110. Ochs, E. (1992) ‘Indexing gender’, in Duranti, A., and Goodwin, C. (eds.) (1992) Rethinking Context: Language as an Interactive Phenomenon, Cambridge, Cambridge University Press, pp. 335–359. (1991) ‘Misunderstanding children’, in Coupland, N., Giles, H., and Wiemann, J. (eds.) ‘Miscommunication’ and Problematic Talk, London, Sage. Okamoto, S. (2000) ‘Rethinking the relationship between language, politeness and gender’, Conference paper to the International Gender and Language Association, Stanford University, California. (1995) ‘“Tasteless” Japanese: less “feminine” speech among young Japanese women’, in Hall, K., and Bucholtz, M. (eds.) Gender Articulated, London and New York, Routledge, pp. 35–51. Paoletti, I. (1998) ‘Handling “incoherence” according to the speakers’ onsight categorisation’, in Antaki, C., and Widdicombe, S. (eds.) Identities in Talk, London, Sage, pp. 170–190. Pavlidou, T. (2000) ‘Telephone conversations in Greek and German: attending to the relationship aspect of communication’, in Spencer-Oatey, H. (ed.) Culturally Speaking: Managing Rapport through Talk Across Cultures, London, Continuum, pp. 121–140. Pearce, L. (1994) Reading Dialogics, London, Edward Arnold. Philips, S. (1992) ‘The routinization of repair in courtroom discourse’, in Duranti, A., and Goodwin, C. (eds.) (1992) Rethinking Context: Language as an Interactive Phenomenon, Cambridge, Cambridge University Press, pp. 311–323. Pilkington, J. (1998) ‘“Don’t try and make out that I’m nice!” The different strategies women and men use when gossiping’, in Coates, J. (ed.) Language and Gender: A Reader, Oxford, Blackwell, pp. 254–269. Pinker, S. (1994) The Language Instinct: The New Science of Language and Mind, Harmondsworth, Penguin. Pomerantz, A. ‘Agreeing and disagreeing with assessments: some features of preferred/dispreferred turn shapes’, in Atkinson, J., and Heritage, J. (eds.) Structures of Social Action, Cambridge, Cambridge University Press.

Bibliography

261

Queen, R. (1997) ‘I don’t speak Spritch: locating lesbian language’, in Livia, A., and Hall, K. (eds.) Queerly Phrased: Language Gender and Sexuality, London, Routledge, pp. 233–242. Reilly, R. (1991) ‘Miscommunication at the person-machine interface’, in Coupland, N., Giles, H., and Wiemann, J. (eds.) ‘Miscommunication’ and Problematic Talk, London, Sage. Rintell, E. (1981) ‘Sociolinguistic variation and pragmatic ability: a look at learners’, in International Journal of the Sociology of Language, 27, pp. 11–34. Robinson, P. (1988) The Women of Botany Bay: A Reinterpretation of the Role of Women in the origins of Australian Society, Ringwood, Vic., Australia, Penguin. Rundquist, S. (1992) ‘Indirectness: a gender study of flouting Grice’s maxims’, in Journal of Pragmatics, 18, pp. 431–449. Sacks, H. (1984) ‘On doing “being ordinary”’, in Atkinson, J., and Heritage, J. (eds.) Structures of Social Action, Cambridge, Cambridge University Press. Safia Mirza, H. (ed.) (1997) Black British Feminism, London, Routledge. Salih, S. (2002) Judith Butler, London, Routledge. Sandoval, C. (1991) ‘US Third world feminism: the theory and method of oppositional consciousness in the post-modern world’, in Genders, no. 10, pp. 1–24. Sargent, L. (ed.) (1981) The Unhappy Marriage of Marxism and Feminism, London, Pluto. Scarcella, R., and Brunak, J. (1981) ‘On speaking politely in a second language’, in International Journal of the Sociology of Language, 27, pp. 59–76. Schaffer, K. (1989) Women and the Bush: Forces of Desire in the Australian Cultural Tradition, Cambridge, Cambridge University Press. Schegloff, E. (1997) ‘Whose text? Whose context?’ in Discourse and Society, 8/2, pp. 165–185. (1992) ‘In another context’, in Duranti, A., and Goodwin, C. (eds.) (1992) Rethinking Context: Language as an Interactive Phenomenon, Cambridge, Cambridge University Press, pp. 191–229. (1984a) ‘On some gestures’ relation to talk’, in Atkinson, J., and Heritage, J. (eds.) Structures of Social Action, Cambridge, Cambridge University Press. (1984b) ‘On some questions and ambiguities in conversation’, in Atkinson, J., and Heritage, J. (eds.) Structures of Social Action, Cambridge, Cambridge University Press. Schiffrin, D. (1990) ‘The management of a co-operative self during argument: the roles of opinions and stories’, in Grimshaw, A. (ed.) Conflict Talk: Sociolinguistic Investigations of Arguments in Conversations, Cambridge, Cambridge University Press.

262

Bibliography

Scollon, R., and Scollon, S. W. (1995) Intercultural Communication: A Discourse Approach, Oxford, Blackwell. Scott, J. (1990) Domination and the Arts of Resistance: Hidden Transcripts, New Haven and London, Yale University Press. Sell, R. (1992) ‘Literary texts and diachronic aspects of politeness’, in Watts, R., Ide, S., and Ehlich, K. (eds.) Politeness in Language: Studies in Its History, Theory and Practice, Berlin and New York, Mouton de Gruyter, pp. 109–130. Shaw, S. (2002) Language and Gender in Political Debates in the House of Commons, Ph.D. thesis, Institute of Education, London University. Sifianou, M. (1992) Politeness Phenomena in England and Greece, Oxford, Clarendon. Skeggs, B. (1997) Formations of Class and Gender: Becoming Respectable, London, Sage. Smith-Hefner, N. (1988) ‘Women and politeness: the Javanese example’, in Language in Society, 17, pp. 535–554. Spencer-Oatey, H. (2000) ‘Rapport Management: a framework for analysis’, in Spencer-Oatey, H. (ed.) (2000) Culturally Speaking: Managing Rapport through Talk Across Cultures, London, Continuum, pp. 11–46. (ed.) (2000) Culturally Speaking: Managing Rapport through Talk Across Cultures, London, Continuum. Spencer-Oatey, H., Ng, P., and Li, D. (2000) ‘Responding to compliments: British and Chinese evaluative judgements’, in Spencer-Oatey, H. (ed.) Culturally Speaking: Managing Rapport through Talk Across Cultures, London, Continuum, pp. 98–120. Spencer-Oatey, H., and Jianyu, X. (2000) ‘A problematic Chinese business visit to Britain: issues of face’, in Spencer-Oatey, H. (ed.) Culturally Speaking: Managing Rapport through Talk Across Cultures, London, Continuum, pp. 272–288. Spender, D. (1980) Man Made Language, London, Routledge. Sperber, D., and Wilson, D. (1986) Relevance: Communication and Cognition, Oxford, Blackwell. Spivak, G. (1990) The Post-colonial Critic: Interviews, Strategies, Dialogues, ed. S. Harasym, London, Routledge. Stalpers, J. (1992) ‘Between matter-of-factness and politeness’, in Watts, R., Ide, S., and Ehlich, K. (eds.) Politeness in Language: Studies in Its History, Theory and Practice, Berlin and New York, Mouton de Gruyter, pp. 210–230. Tanaka, N., Spencer-Oatey, H., and Cray, H. (2000) ‘“It’s not my fault!” Japanese and English responses to unfounded accusations’, in Spencer-Oatey, H. (ed.) Culturally Speaking: Managing Rapport through Talk Across Cultures, London, Continuum, pp. 75–97. Tannen, D. (ed.) (1993) Gender and Conversational Interaction, Oxford and New York: Oxford University Press.

Bibliography

263

(1991) You Just Don’t Understand: Women and Men in Conversation, London, Virago. (1990) ‘Silence as conflict management in fiction and drama’, in Grimshaw, A. (ed.) Conflict Talk: Sociolinguistic Investigations of Arguments in Conversations, Cambridge, Cambridge University Press. Tannen, D., and Saville-Troike, M. (eds.) (1985) Perspectives on Silence, Norwood, Ablex. Taylor, T. (1992) Mutual Misunderstanding: Scepticism and the Theorising of Language and Interpretation, London, Routledge. Thompson, E. P. (1980/1st pub.1963) The Making of the English Working Class, London, Penguin. Thornborrow, J. (2002) Power Talk: Language and Interaction in Institutional Discourse, Harlow, Longman. Toolan, M. (1996) Total Speech: An Integrational Linguistic Approach to Language, Durham and London, Duke University Press. Troemel-Ploetz, S. (1998) ‘Selling the apolitical’, in Coates, J. (ed.) Language and Gender: A Reader, Oxford, Blackwell, pp. 446–458. Trudgill, P. (1972) ‘Sex, covert prestige and linguistic change in the urban British dialect of Norwich’, in Language in Society, 1, pp. 179–95. Trudgill, P., and Chambers J. (eds.) (1991) Dialects of English, Harlow, Longman. Volosinov, V. (1973) Marxism and the Philosophy of Language, trans. Matejka, L. and Titunik, I., New York, Seminar Press. Vuchinich, S. (1990) ‘The sequential organisation of closing in verbal family conflict’, in Grimshaw, A. (ed.) Conflict Talk: Sociolinguistic Investigations of Arguments in Conversations, Cambridge, Cambridge University Press. Walper, S., and Valtin, R. (1992) ‘Children’s understanding of white lies’, in Watts, R., Ide, S., and Ehlich, K. (eds.) Politeness in Language: Studies in its History, Theory and Practice, Berlin and New York, Mouton de Gruyter, pp. 231–254. Walsh, C. (2001) Gender and Discourse: Language and Power in Politics, the Church and Organisations, Harlow, Longman Pearson. Walters, J. (1981) ‘Variation in the requesting behaviour of bilingual children’, in International Journal of the Sociology of Language, 27, pp. 77–92. Wareing, S. (1995) ‘Gender, Speech Styles and the Assessment of Discussion’, unpublished Ph.D. thesis, Glasgow, Strathclyde University. Watts, R. (1992a) ‘Linguistic politeness and politic verbal behaviour’, in Watts, R., Ide, S., and Ehlich, K. (eds.) Politeness in Language: Studies in its History, Theory and Practice, Berlin and New York, Mouton de Gruyter, pp. 43–70. (1992b) ‘Acquiring status in conversation: “male” and “female” discourse strategies’, in Journal of Pragmatics, 18/5, pp. 467–505.

264

Bibliography

Watts, R., Ide, S., and Ehlich, K. (1992) ‘Introduction’, in Watts, R., Ide, S., and Ehlich, K. (eds.) Politeness in Language: Studies in its History, Theory and Practice, Berlin and New York, Mouton de Gruyter, pp. 1–21. Webster, W. (1990) Not a Man to Match Her, London, Women’s Press. Wenger, E. (1998) Communities of Practice, Cambridge, Cambridge University Press. Werkhofer, K. (1992) ‘Traditional and model view: the social constitution of the power of politeness’, in Watts, R., Ide, S., and Ehlich, K. (eds.) Politeness in Language: Studies in its History, Theory and Practice, Berlin and New York, Mouton de Gruyter, pp. 155–199. Werth, P. (1999) Text Worlds: Representing Conceptual Space in Discourse, Harlow, Longman. West, C., and Frankel, F. (1991) ‘Miscommunication in medicine’, in Coupland, N., Giles, H., and Wiemann, J. (eds.) ‘Miscommunication’ and Problematic Talk, London, Sage. Whelehan, I. (1999) Overloaded: Popular Culture and the Future of Feminism, London, Women’s Press. Widdicombe, S. (1998a) ‘But you don’t class yourself: the interactional management of category membership and non-membership’, in Antaki, C., and Widdicombe, S. (eds.) Identities in Talk, London, Sage, pp. 52–71. (1998b) ‘Identity as an analyst’s and a participant’s resource’, in Antaki, C., and Widdicombe, S. (eds.) Identities in Talk, London, Sage, pp. 191–206. Wierzbicka, A. (1999) Emotions Across Languages and Cultures: Diversity and Universals, Cambridge, Cambridge University Press. Wilkinson, S., and Kitzinger, C. (eds.) (1993) Heterosexuality: A Feminism and Psychology Reader, London, Sage. Wittig, M. (1992) The Straight Mind and Other Essays, Hemel Hempstead, Harvester. Wodak, R. (2001) ‘What CDA is about – a summary of its history, important concepts and its development’, in Wodak, R., and Meyer, M. (eds.) Methods of Critical Discourse Analysis, London, Sage. (1998) Disorders of Discourse, Longman, Harlow. Wodak, R., and Meyer, M. (eds.) (2001) Methods of Critical Discourse Analysis, London, Sage. Wooffitt, R., and Clark, C. (1998) ‘Mobilising discourse and asocial identities in knowledge talk’, in Antaki, C., and Widdicombe, S. (eds.) Identities in Talk, London, Sage, pp. 107–121. Yedes, J. (1996) ‘Playful teasing: kiddin’ on the square’, Discourse and Society, 7/3, pp. 417–438. Ylanne-McEwen, V., and Coupland, N. (2000) ‘Accommodation theory: a conceptual resource for intercultural sociolinguistics’, in SpencerOatey, H. (ed.) Culturally Speaking: Managing Rapport through Talk Across Cultures, London, Continuum, pp. 191–216.

Bibliography

265

Yi, L. (2000) ‘What motivates compliments and how they function: a pragmatic study of compliments in relation to face-effect in a crosscultural academic context’, discussion paper, Nanking University. Zegarac, V., and Pennington, M. (2000) ‘Pragmatic transfer in intercultural communication’, in Spencer-Oatey, H. (ed.) Culturally Speaking: Managing Rapport through Talk Across Cultures, London, Continuum, pp. 165–190. Zimin, S. (1981) ‘Sex and politeness: factors in first and second language use’, in International Journal of the Sociology of Language, 27, pp. 35–58. Zimmerman, D. (1998) ‘Identity, context and interaction’, in Antaki, C., and Widdicombe, S. (eds.) Identities in Talk, London, Sage, pp. 87–106.

Index

Adams, K., 192 African American, 190 Afshar, H. and Maynard, M., 182 aggression, 65, 156, 191, 193 Allbright, M., 191, 193 Althusser, L., 26 anecdotes, 12, 44, 115, 166 Antaki, C. and Widdicombe, S., 19, 20, 50 anti-essentialism, 241 apology, 61, 111, 112, 130, 221–226, 229, 234 appropriateness, 4, 9, 10, 13, 70–71, 110, 144, 194, 226, 246 Ariel, M. and Giora, R., 180 army, 126 assertive, 168 Atkinson, J. and Heritage, J., 21, 45 Azim, F., 18 background knowledge, 28 Baker, S., 25 banter, 108, 124, 156, 158 Baran, D. and Syska, O., 191 Bargiela, F., 77, 105 Bargiela, F. et al., 70, 145 Beaken, M., 33 Beebe, L., 115, 125, 132, 133, 136 Bell, D. et al., 186 Bem, S., 171 Bergvall, V. et al. (eds.), 2, 170, 216 Berk-Seligson, S., 64 Bhabha, H., 241 binary thinking, 169–174 Bing, J. and Bergvall, J., 168, 170–171 Blum-Kulka, S., 44, 69, 74, 75, 140 Bolton, M. and Bolton, J., 151–152, 205, 206, 207 Bourdieu, P., 3, 35, 36, 47, 71, 184 Boz, C., 105

266

Braid, M., 194 Brayford, C., 7 Brown, P., 207–213, 229, 230 Brown, P. and Levinson, S., 5, 12, 13, 32, 33, 41, 122, 134, 178, 212–213, 219, 244, 245 Bucholtz, M., 4, 30, 70, 182, 190, 196 Burman, E. and Parker, I., 242, 244 Butler, J., 4–5, 23–24, 26, 47, 100, 170, 173 Cameron, D., 2, 43, 115, 155, 166, 167, 168, 180, 185, 187, 190, 212 Campaign for Courtesy, 6, 206 Chan, G., 169, 181 Chomsky, N., 24 Christie, C., 2, 67 class, 20, 33, 47, 48, 63, 64, 65, 106, 149–153, 174, 181–184, 204 Coates, J., 166, 167, 175, 179, 188, 213 Coates, J. and Cameron, D., 44, 170, 183 co-construction, 28 compliments, 219–221 conflict, 168, 187 Co-operative Principle, 37–38, 61 co-operative skills, 167, 179 community of practice, 2, 3, 4, 5, 9, 10, 30, 98, 109, 126, 135, 154, 159, 194, 195–197, 239 context, 29, 48–49, 67, 144, 174, 194, 235 contextual, 2 contextualisation cue, 39 Conversation Analysis/CA, 17, 46, 108, 238, 242, 243, 244 Coulmas, F., 68 Coupland, N., 6, 9 courtesy, 7, 63, 204, 205–207

Index Coward, R., 240 Cox, S., 191, 192 Crawford, M., 5, 170, 173 Crawford, M. and Chaffin, R., 186 Critical Discourse Analysis/CDA, 17, 46, 49, 108, 238, 242, 243, 244 cross-cultural analysis, 31, 105 Cross-Cultural Linguistic Politeness Research Group, viii, 245 Culpeper, J., 83, 122, 123, 124, 125, 136, 146 culture, 30–32, 105, 145 Daley, J., 7 Dalrymple, 7 Data, 10, 13, 43, 45, 46–48, 115 Davidoff, L. and Hall, C., 183 deference, 2, 7, 105, 164, 180, 190, 208, 211 determinism, 35 deviation, 81 Diamond, J., 5, 84, 100, 101, 240 difference, 164–169, 224 directness, 134, 139, 141, 142, 143, 149, 165, 172, 184, 188, 204, 233 discourse, 3, 5, 25, 82 discursive structures, 190 distance, 100, 101 dominance, 164–169 dyad, 25 Eckert, P., 29 Eckert, P. and McConnell-Ginet, S., 3, 4, 30, 195, 196 Edley, N. and Wetherall, M., 188 Edwards, V., 182 education, 20 Ehlich, K., 64, 66, 72 Eelen, G., 3, 4, 8, 12, 31, 35, 36–37, 43, 57, 71, 74, 106, 121, 124, 133, 140, 149, 151–152 Erlich, S., 140 Erlich, S. and King, R., 140 essentialism, 170, 238 etiquette, 63, 66, 151–152, 204–205, 206 ethnomethodology, 19, 46 eye movement, 45 face, 41, 58, 77–79 Face Threatening Act /FTA, 6, 58, 60, 76, 78, 79, 81, 88, 127, 211 Fairclough, N., 28, 192, 242 feminine, 156, 157, 158, 159, 167, 175, 184, 210, 227 femininity, 165, 184–190, 203, 204

267 feminist linguistics, 164, 195, 238 feminist theory, 164, 185 first-order politeness, 8 Foley, W., 31 Foucault, M., 3, 19, 24–25, 100, 174 Frankenberg, R., 182 Fraser, B. and Nolen, W., 134, 144 Freed, A., 5, 173, 194 Fukushima, S., 32, 57, 68, 77, 101 Fuss, D., 170 gay speech styles, 171 gender, 4–5, 155, 158, 164–198, 202–235, 239 gender-appropriate, 223, 225 gendered domains, 5 gendered identity, 9 gendered practices, 5 gendered stereotypes, 164, 184–190 generalisations, 215 Gilbert, S. and Gubar, S., 186 Goffman, E., 58 Goodwin, C., 21, 28, 45, 48 Goodwin, C. and Harness Goodwin, M., 12 Grainger, K., 73 Gregory, I., 7, 206 Grice, H. P., 36–37, 38, 61, 80 Grimshaw, A. (ed.), 6, 9 Gumperz, J., 20, 39 Gunthner, S., 143–144 habitual style, 106–107 habitus, 3, 35, 36, 48, 71, 184 Halberstam, J., 187, 190 Hall, S., 34 Hamza, A., 32, 84 Harness Goodwin, M., 150 Harris, S., 77, 127, 141 Harris-Bond, M. et al., 67 Haugen, J., 191 Haverkate, H., 92, 122, 135 Held, G., 66, 77, 116 Henley, N., 150, 173 heterosexuality, 182 Hochschild, 205 Holland, S., 188, 241 Holmes, J., 2, 6, 33, 43, 59, 73, 83, 109, 167, 179, 197, 207, 209, 213, 229 Holmes, J. and Meyerhoff, M., 30 honorifics, 68 House, J., 143, 147–148 House of Commons, 127 humour, 73, 156, 233

268 hypercorrectness, 208 hypothesising, 71 hypothesised appropriateness, 110 hypothesised intention, 136 Ide, S. et al., 140, 145 identity, 20 impolite, 6, 13, 88 impoliteness, 121–159 indirectness, 14, 75, 130, 142, 165 individual, 17, 18, 19, 29–33 inference, 84, 88 instigating, 150 institutions, 197 institutional history, 24 institutional status, 100 insult, 126 integrational linguistics, 2, 29, 34, 244 interactional history, 22 interactional power, 157, 175, 240 interpretation, 16–51 intention, 11, 22–24, 91 Jackson, P., 190 Janney, R. and Arndt, H., 67, 70, 90 Jarrett-Macaulay, D., 182 Jary, M., 62, 80, 102, 135, 144, 145 Johnson, S. and Meinhof, U., 170 Jucker, A., 64 judgement, 8, 33, 72–73, 110, 112–114, 139–149, 152 Kaspar, G., 116 Keenan, E., 208 Keinpointer, M., 116, 121, 123, 124, 125, 146 Kendall, G. and Wickham, G., 16 Kharraki, A., 226 Klerk De, V., 188 Kristeva, J., 18 Labov, W., 127, 149 ladette, 191, 192 Lakoff, R., 1, 2, 61, 165, 174, 179, 184, 197, 203 Leap, W., 172 Leech, G., 61 lesbian speech styles, 171–173 Lewis, R. and Mills, S., 182 Liladhar, J., 19, 48, 154, 186 literary theory, 16, 19 Livia, A. and Hall, K., 172 local rank, 100 Lycan, W., 123

Index Manke, M., 5, 130, 131–132 manners, 203 Mao, L., 105 masculinity, 142, 159, 168, 175, 184–190, 204 Matsuda, M., 127 Matsumoto, 136 Maynard, M. and Purvis, J., 182 McClintock, A., 46, 174, 181 McElhinney, B., 5, 194 meta-discourse, 82 methodology, 10, 89 Meyer, M., 242 Miller, L., 141, 143 Mills, S., 2, 3, 21, 22, 34, 50, 149, 174, 175, 184, 190 Minh-ha, T., 182 Minsky, 25 miscommunication, 6 misunderstanding, 6, 9, 41, 42, 98, 114, 231 model of communication, 33–43 Model Speaker, 17–29, 89, 91 Modleski, T., 241 Mohanty, C., 182 Montgomery, M., 60 Morgan, M., 150, 173 Mowlam, M., 191, 193 Moyles, C., 192 mutual manifestness, 28, 114 negative face, 78 negative politeness, 32, 59, 75, 76, 105, 210, 230 New Criticism, 11 new lad, 189 new man, 188, 189 Nichols, P., 182 O’Barr, W. and Atkins, B., 64, 179, 180 Ochs, E., 239 Okamoto, S., 145, 191 Other, 26, 27 over-politeness, 225 Paoletti, I., 73 Pavlidou, T., 143 Pearce, L., 27 performativity, 5, 13, 173 Pilkington, J., 188 Pinker, S., 34 playful teasing, 123 politeness, 57–116, 121–159, 202–235

Index politeness1, 8 politeness2, 8 Politeness Principle, 61 Polite Society, the, 6 politic behaviour, 68 positive face, 78, 137 positive politeness, 2, 32, 59, 75, 76, 105, 156, 205, 210, 214, 215, 229 post-colonial theory, 16 power, 5, 48, 84, 100–101, 166, 239, 240 powerlessness, 174, 179, 190, 203, 235 power relation, 177 practical rationality, 91, 111 pragmatics, 2, 6, 8, 18, 43 private sphere, 73, 187 process, 38 professional, 194 proxemics, 155 psychoanalytical theory, 18 public sphere, 73, 168, 190 ‘punctual’ analysis, 170 qualitative analysis, 43 quantitative analysis, 43, 214 Queen, R., 172 Queer theory, 171 Questionnaires, 44 race, 20, 33, 46, 48, 149–153, 174, 181–184 racism, 24, 243 Rank of imposition, 100 rapport management, 78 rapport talk, 166 reformulation, 97 Relevance theory, 5, 62 report talk, 166 respectable, 183 ritual, 128 Robinson, A., 193 Safia-Mirza, H., 182 Salih, S., 173 Sandoval, C., 182 Sapir-Whorf model, 34 Schegloff, E., 17, 20, 47, 49 Schiffrin, D., 12 school, 128 Scollon, R. and Scollon, S., 17, 27, 30, 31, 37, 38, 41, 75 Scott, J., 106 second-order politeness, 8 self, 18, 19, 27

269 Sell, R., 59, 63 sexism, 24, 243 Shaw, S., 17, 67, 127, 168 Sifianou, M., 37, 57, 60, 65, 75, 105 sincerity, 60, 144 Skeggs, B., 19, 152, 183–184, 186, 193 Smith-Lovin and Brody, 169 Social politeness, 67, 70 Smith-Hefner, N., 212 Speech Act theory, 38, 80, 81 Spencer-Oatey, H., 44, 78, 112, 121, 148 Spencer Oatey, H. and Jianyu, X., 137–139 Spender, D., 2, 165, 174 Sperber, D. and Wilson, D., 2, 28–29, 62–63, 92 Stalpers, J., 141 Stearns, 189 Stereotype, 1, 4, 13, 32, 73, 142, 148, 152, 157, 166, 172, 174, 181, 183, 184–188, 196, 202–207, 210, 214, 218, 223, 227, 238, 241, 246 straight speech, 172 strategies, 74 strengthening particles, 209 strong women speakers, 164, 190–195 structural inequalities, 181 swearing, 108, 139, 140, 149, 172, 184, 188, 193, 204 tact, 68 tag-questions, 216 Tannen, D., 2, 166, 175, 179, 188, 213 Tannen, D. and Saville-Troika, M., 99 Taylor, T., 82–83, 111 telementation, 34 Thatcher, M., 191 Third-Wave Feminist Linguistics, 2, 239, 240 Thompson, E. P., 47 Thornborrow, J., 5, 17, 97, 125, 174, 181, 242, 243 transcription, 45 Toolan, M., 2, 21–22, 23, 33, 38, 40, 45, 46, 48, 51, 92, 110, 111, 114, 116 Troemel-Ploetz, S., 166 Trudgill, P., 208 Trudgill, P. and Chambers, J., 150 unconscious, 18, 19 understanding, 40 universality, 105

270 Van Dijk, T., 242 variable, 1, 46, 100–104, 164, 174–184, 213 Volosinov, V., 26–27 Walsh, C., 48, 60, 70, 154, 157, 167, 192, 193 Watts R. et al. (eds.), 8, 57, 63, 66, 68, 69 Webster, W., 191 Wenger, E., 3, 30, 195, 196 Werkhofer, K., 18, 64, 65, 76, 91, 97, 103 Werth, P., 28

Index Wareing, S., 167 Whelehan, I., 189, 191 whiteness, 182 Wierzbicka, A., 11 Wilkinson, S. and Kitzinger, C., 182 Wittig, M., 172 Wodak, R., 9, 13, 41, 88 Wodak, R. and Meyer, M., 242 ‘Women’s language’, 2, 164, 165, 179, 196, 203, 207, 224 Yedes, J., 124, 154 Zimmerman, D. and West, C., 169

Suggest Documents